The Status of Psychosocial Research in the Rehabilitation the Cancer Patient

of

Ivan Barofsky

N

OVEMBER, 1991, marked the twentieth anniversary of passageof the National Cancer Act and the initiation of the “War on Cancer.” One commentator describes the results to date as “a spectacular failure,“’ although the basic researchinitiated by theseefforts was acknowledged as creating the potential for future accomplishments. A somewhat different assessmentmay be made for the cancer rehabilitation program sponsored by the National Cancer Institute; minimal research accomplishmentswith little or no stimulus for expansion of the scientific basis for cancer rehabilitation, but a dramatic improvement in available cancer rehabilitation services.2,3Thus, although there is now a relatively firm scientific foundation for cancer prevention and treatment, the samecannot be said for cancer rehabilitation. Understanding the events that led to this outcome is important, not only to properly grasp the current status of psychosocial aspects of cancer rehabilitation, but also for setting the stagefor future research efforts. What will be learned is that which is known about the psychosocial aspectsof cancer rehabilitation; it is a complex product of investigator interests and efforts at directing scientific activities by various research-sponsoring agencies.This article will describe how the development of this researcharea was confounded leaving the intrinsically interesting questionsraised by the field essentially unanswered. In addition, examples will be given of different psychosocial researchin the area of cancer rehabilitation. Finally, From the Department of Psychiatry and Behavioral Science, The Johns Hopkins University School of Medicine, Baltimore, MD. This article is partially based on the author’s experience as Program Director of Continuing Care, Community Oncology and Rehabilitation Branch, Division of Cancer Control and Prevention, National Cancer Institute from October, 1982, to October, 1984. The opinions stated are his own and they are not to be construed as in any way reflecting the views of the National Cancer Institute. Address reprint requests to Ivan Barofsky, PhD, Division of Digestive Diseases, Francis Scott Key Medical Center, 4940 Eastern Ave, Baltimore, MD 21224. Copyright 6 1992 by W.B. Saunders Company

0749-2081/92/0803-OGO6$5.OOlO 190

a seriesof specific recommendationswill be made for future research activities that are designed to encouragethe rational development of psychosocial aspectsof cancer rehabilitation. HISTORICAL PERSPECTIVES

The first, and critical, observation to be made is that cancer rehabilitation was not included as an explicit objective of the National Cancer Act of 1971.4 It was included as one of the seven objectives of the National Cancer Program, which was to implement the National Cancer Act. Its inclusion in the Program was justified by a restricted interpretation of the mandatethat was statedin the National Cancer Act of 1971 for reducing cancer morbidity. A Planning Conferenceto develop this Program, chairedby John Healy and Guy Robbins, was held December4 to 8, 1972, and culminated in the publication of the Rehabilitation Planning Report of June, 1973.5 Cancer rehabilitation was finally included as a specific legislated objective following the amendmentsof the National Cancer Act passedby Congressin 1978.6 This objective was further reinforced by the National Institutes of Health Extension Act of 1985.7 Clearly, there has been an intent on the part of Congress to support rehabilitation of the cancer patient, but historically and programmatically it is important to note that such support has been in responseto a continuing political processthat has pitted major players againsteachother; eachstruggling to set researchpriorities and resourceallocations for this new national commitment. For example, when Congress and President Nixon proposedthe “War on Cancer,” the National Cancer Institute was just one of severalof the National Institutes of Health and was intellectually dominated by a biomedical model of cancer that emphasized cure over prevention and biology over behavior. Political support for cancerrehabilitation camefrom practicing oncologists who were aware of the rehabilitative needsof their patients, but also becausea variety of well-known personalities of that time developedcancer which resulted in a significant increase in media coverage of the treatment and rehabilitation of cancer. Still, rehabilitaSeminars in Oncology Nursing, Vol 8, No 3 (August). 1992: pp 190-201

PSYCHOSOCIAL

RESEARCH

AND REHABILITATION

tion was not thought of as generating basic research questions that justified support in the samesensethat viral biology did. Part of the reason for this can be found in the very definition of what is meant by rehabilitation. Gunn’ defines rehabilitation as “the restoration of a patient with residual deficits as a result of his diseaseor treatment to as normal a functional state as possible.” Thus, rehabilitation is always playing catch-up, trying to help restore what has been lost. For this reason alone, rehabilitation is relegateda secondaryrole in the hierarchy of activities involved in the medical managementof the cancer patient, with prevention of diseaseonset and combating a diseaseprocessgiven primary roles. Even the argument that maximum functional outcome comes from integrating the rehabilitative needsof a patient into the design of a treatment does not avoid the reality that rehabilitation-related issues only state what not to do, not what to do to avoid the deficits created by a diseaseor its treatment. Somehave argued(eg, Gotay, personalcommunication, 1992) that it is not reasonableto expect true restoration of function following cancer treatment. Rather, it is more likely that a person be restored to a “maximum” level of functioning. Although this definition may reflect what happens in practice, it can also be seen as placing the responsibility of rehabilitation on the patient. This may be considered unfair considering the adverse consequenceof cancer and its treatment. If Gunn’s* definition of rehabilitation is considered an ideal and the second is what occurs in practice, then clearly a dynamic has to exist between them so that practice approachesthe ideal. However, neither definition deals with the fact that rehabilitation has to be justified by more then reference to humane treatment of patients, or simply that patients want such services. If cancer rehabilitation is to have equal footing in the political process that leads to researchpromotion and accomplishments, then its basic researchneedshave to be clearly articulated. This is also true for psychosocial aspectsof cancer rehabilitation. The Rehabilitation Planning Report’ proposed support for three objectives: 1. To develop the capacity to provide rehabilitation to all cancer patients. 2. To develop the means to restore maximum physical and functional capabilities to cancer patients.

191

3. To develop the means to restore cancer patients’ life-style and reinforce re-entry into the community. The third objective was further delineated to include comprehensive rehabilitation services and techniques, psychosocial rehabilitation, and vocational rehabilitation. The specific tasks for this objective are described in Table 1. Burke (personal communication, 1982) describes implementation of these Program objectives, from 1973 to 1979, as including contract support for 55 projects in 13 specific areas and grants for 46 projects, 39% of which were for psychosocial research. The contract mechanism was used by the National Cancer Institute to promote implementation of known principles and techniques. Support was provided for professional education programs, demonstration programs, etc., designedto increasethe proportion of patients who received rehabilitative services. Rehabilitation research was to be supported by investigator-initiated proposals (ie, the “ROl” or “R18” mechanisms) with minimal programmatic guidance from the National CancerInstitute. The original form of the Rehabilitation Program remained intact until 1979, when a reorganization occurred and the program was folded into a behavioral medicine program. Although rehabilitation researchhas continued to be supported by the National Cancer Institute, it has never regained the level or extent Table 1. Develop Patient

improved

Means

Life-Style and Reinforce the Community

to Restore Re-Entry

Cancer Into

Approach NO.

Comprehensive 7.3.1.1 7.3.1.2

Psychosocial 7.3.2.1 7.3.2.2 7.3.2.3 7.3.2.4 7.3.2.5 7.3.2.6

Vocational 7.3.3.1 7.3.3.2 7.3.3.3 7.3.3.4

rehabilitation services and techniques Multidisciplinary rehabilitation concepts Integrated rehabilitation management techniques rehabilitation Patient information and counseling Family information and counseling Professional attitudes and motivation Public attitudes and motivation Psychosocial counseling procedures and techniques Lay volunteer recruitment and training

rehabilitation Vocational opportunities and counseling Vocational training and retraining Employment opportunities Employment conditions

Data from the Rehabilitation

Planning

Report5

192

of commitment that was true during the initial phasesof the “War on Cancer.” Evidence for this includes the systematic administrative downgrading of interest in rehabilitation (from a program to a branch), and the only occasionalinitiation of new rehabilitation researchprograms in the last 8 to 10 years. The rapid ascendanceand decay of the Rehabilitation Program was becauseof many reasons,but probably most important was the mistaken assumption that enough was known about the rehabilitation of the cancer patient so that complex multi-million dollar programs could be initiated with confidence, and that investigators could proposeresearchthat had a reasonablechanceof generating useful information. In addition, factors that influence any sponsored-research activity (eg, changesin the administrative headof the program, the personalities of the staff, and so on) also affected the Rehabilitation Program. As Gotay and Yates2state, many of the contracts were not properly evaluated so that they had limited generalizability, whereas much of the sponsoredresearch also had problems of imprecise definitions, inadequate outcome measures,and limited possibilities of generalizability. However, of the many programs initiated during this period, including psychosocial initiatives, some still persist and can be considered classic in conceptualization and implementation (see below). Table 2 summarizes the author’s view of some of the accomplishmentsof the Rehabilitation Programfrom 1973to 1983. Table 2 should not be read as contradicting the Gotay and Yates2 study, which is partially basedon the samedata. Rather it should be read as illustrating that something can be accomplishedeven when a program is a frank failure, prematurely terminated, and so on. However, as a consequenceof program deficiencies, the Rehabilitation Program was not able to sustain the institutional support within the National Cancer Institute neededto overcome the difficulties inherent to the start-up of any sponsored researchprogram. A snapshotof current National Cancer Institute support for cancerrehabilitation can be shown by a recent review of 51 cancer control protocols that are sponsoredby the researchbasesfor the Community Clinical Oncology Program.’ Of the 51 protocols, 13.7% dealt with pain control (all of which are drug-orientated) and 11.8% dealt with what can be broadly defined as psychosocialissues

IVAN BAROFSKY

Table 2. Accomplishments of the Cancer Rehabilitation Program-1973-1983 1. The emotional and psychosocial natural history of cancer and its treatment endure through stages, over time, much as any developmental process. 2. Early intervention modulates the emotional and psychosocial sequelae associated with cancer and its treatment. 3. Integrating treatment and rehabilitation minimizes the adverse consequences of cancer and its treatment. dynamics and principles of behavior 4. Organizational change will determine which cancer rehabilitation programs will succeed. can restore function in cancer 5. Physical rehabilitation patients. affect 6. Support programs for cancer patients positively indicators of well-being. 7. Training programs in the absence of adequate prior planning for integrating practitioners into the cancer care system will probably not succeed. 8. Mainstreaming insures optimal adaptation and coping by pediatric cancer patients. 9. Hospice research has inherent limitations. are more 10. Team approaches to cancer rehabilitation likely to succeed than non-team approaches. groups can effect the psychosocial 11. Community response to cancer and its treatment. 12. It is possible to control the pain of the cancer patients. This table is based on the author’s review of the research sponsored by the Rehabilitation Program of the National Cancer Institute during the period of 1973-1983.

(eg, coping with treatment side effects). There were no studies of physical restoration, even though physical restoration has been a particularly successful area of past rehabilitation research (Burke, personal communication, 1982). What is impressive about this outcome is that it occurred at such a high frequency, considering the emphasisin the Community Clinical Oncology Programon primary and secondary prevention. What it demonstratesis that, when given an opportunity, patientrelated issues, such as rehabilitation, continue to be of interest to oncological researchers. Another perspective on current federal support for cancer rehabilitation can be obtained from the “Report of the Task Force on Medical Rehabilitation Research.“lo This report was prepared in responseto the formation of the “National Center for Medical Rehabilitation Research.” The center was designed to stimulate information dispersal and research.It again acknowledgesthe continued clinical and political relevanceof rehabilitation efforts, but by isolating rehabilitation from the continuum of treatmentresponsibilities within eachin-

PSYCHOSOCIAL RESEARCH AND REHABILITATION

stitute may actually worsen the problem of integrating rehabilitation into the practice of medicine. Fortunately, a variety of nonfederal organizations, such as the American Cancer Society, have supported rehabilitation services and research,so that the adverseimpact of oscillations in federal support have been blunted. The paradox of this history is that as cancer is becoming a more chronic, less often fatal disease, the proportion of diagnosed cancer patients who would benefit from rehabilitative efforts can be expected to increase,’ ’ yet the magnitude of the problem remains essentially unknown. Although there have been someefforts to determine the service needs of cancer patients,12 none have provided a population-basedestimateof the full range of rehabilitative services required by cancer patients. The researchof Houts et ali3 comesclosest to generating the type of data required for designing a rational rehabilitation research program. Houts et al’s data are based on accruing patients through a state cancer registry. As a result, they have a good chance of identifying the entire population of cancer patients in a specific locale, for some specific time period. One limitation in this type of accrual is the delay between diagnosis, treatment, and registration. However, this is a technical but solvable problem. With populationbasedmeasures,it would be possible to accurately estimate the incidence and prevalence of cancer rehabilitation problems, evaluate interventions, and monitor the impact of interventions on the duration and severity of disability. These types of data are needed if there is to be a firm scientific foundation that would justify prolonged support for cancer rehabilitation studies. The absenceof such data can be traced to an early decision in the development of the Rehabilitation Program to emphasizesocial-impact research(eg, return to work; Burke, personal communication, 1982), as opposed to estimating the magnitude of the problem and guide resourceson the basisof theseestimates. PSYCHOSOCIAL ASPECTS OF CANCER REHABILITATION

Although the literature on psychosocial aspects of cancer rehabilitation continues to expand, the definition of psychosocial remains obscure. To get some senseof what is meant by “psychosocial,” the researchpublished in the Journal of Psychoso-

193

cial Oncology was reviewed by title. What became clear was that a wide range of topics were published under the guise of psychosocial, including symptom management, quality-of-life research, health servicesresearch,etc. Although all of these researchareasare legitimate aspectsof a person’s psychological and social responseto cancer and its treatment, they are basically ancillary to the core notion of what “psychosocial” means; which in this article refers to the capacity of humans to adaptto physical, emotional, and financial assaults and traumas. “Coping, adjustment, adaptation” are different namesfor the sameprocessactivated as a result of a psychosocial responseto a disease and its treatment. Still, even theseprocessescan be studied from several perspectives.The current debate over whether psychooncology is the same as psychosocial oncology is part of the same definitional struggle.l4 In addition, how do studies of concrete needs fit into what is meant by psychosocial’?Is pain part of what is meant by psychosocial? The answers to these questions seem to be there is more than one intellectual tradition mixed together under the rubric of psychosocialresearch, including psychological and psychiatric, sociological, physiological, and economic. Each tradition approachesthe capacity of a person to adapt to cancer from a different perspective. Psychooncology evolved out of the psychiatric literature with its implicit assumptionsconcerning the psychopathological basis of current behavior, and its emphasis on premorbid psychiatric determinants of cancer etiology and disease course.l4 Outcome is measured in terms of psychiatric symptomology, and servicesare provided most often on a consult-liaison basis. Depressionand anxiety are common outcome measures. Social oncology evolved out of the social scienceliterature and includes those elementsof both psychology and economicsthat deal with judgment and decision making, as well as a major input from the sociological literature. Outcome is measuredin terms of quality of life but can also be measuredin terms of behavior, patient, or clinician. Services are often provided by a multidisciplinary team. In social oncology, the social context determines the problems of interest. Selected domains of quality of life, such as work status, are common outcome measures. A third area of psychosocial research, psychophysiological oncology, deals with physiological

194

processesand their psychological and social consequences.Anticipatory nauseaor vomiting, disrupted sexual functioning, pain, and mucositis exemplify the subject matter of interest. Outcomes can be measuredin terms of psychiatric symptomology or quality of life but also in terms of persistence of physical symptoms. Services can be provided on a consult or a multidisciplinary basis. Changes in chronic pain behavior, as a result of cancer treatment, would be a prototypical outcome measure. Need assessmentsof cancerpatients measurethe demand side of the supply and demand equation essential for resource and service allocation. Need assessmentsare inherently reactive, reflecting patient’s perceived needs, and therefore can either overestimate or underestimate true needs. Financial issues, home health care, insurance, or transportation are examples of concrete needs. Most surveys of service needs combine concrete and nonconcrete(eg, emotional) needs.An implicit assumption in this area of researchis that psychosocial issuesevolve from concrete needs, and that if these needs are met (eg, ensuring a person has transportation to a health facility), psychosocial issues would not develop (eg, anxiety about getting to a doctors appointments). With such a broad spectrum of research fitting under the banner of psychosocial research, it is sometimes difficult for the researcher to consistently apply the methodologies of a particular research tradition to a particular research problem. As a result, it is not always possible to determine the exact research question being asked and answered. Thus, both psychooncologist and social oncologist could addressthe issue of social support, but would do so with very different assumptions concerning what is being studied, research strategies, and so on. Too often these conceptual and methodological differences are ignored with an inevitable consequenceon the precision of the researchoutput. How these differences influenced the early Rehabilitation Program is hard to document, but they can be suspectedof playing a role. The social oncologist and psychooncologist also differ in that the social oncologist is more able to approach the study of the psychosocial consequencesof cancer and its treatment independentof the need to justify particular medical or psychological interventions. The reasonfor this is that the psychooncologist usually has a dual role, clinician

IVAN BAROFSKY

and researcher, whereas the social oncologist is usually not a clinician or involved in policy decisions affecting the patient. Thus, involving the social scientist helps ensure that studies of the psychosocial aspects of cancer rehabilitation are unbiasedor reflect an alternative bias. An example of how a social science investigator can raise interesting questions about a well-established clinical observationis shown in the studiesof Cassileth et a1r5*16 showing that psychosocial variables may not be associatedwith length of cancerpatient survival, as popularly assumed. EXAMPLES OF PSYCHOSOCIAL RESEARCH IN ONCOLOGY

Of the many possible examplesof psychooncological research, Project OmegaI is representative. The purposeof the researchwas describedas follows: Although an occasional patient might have found an extended psychiatric contact useful, no one developed u typicalpsychiatric syndrome as a result of cancer, either at the initial impact or during follow-up periods. Our fask, however, was not to discover psychiatric cases. It was to find out how the ‘average’ newly diagnosed cancer patient would cope with common psychosocial problems (p 37). I7

This statement is quite informative because it acknowledged that the cancer patients they interviewed were not psychiatrically disturbed but, as was found, differed in terms of their capacity to cope. Becausethe patients’ capacity to cope was also related to their initial level of distress, a basis was available to justify screeningnew patients and to intervene to prevent the development of psychiatric symptomology. l8 The results of their research suggestthat they were able to significantly reduce the emotional distressof the patients who received an intervention (four sessionsof either consultation-problem solving or cognitive therapy) relative to controls who did not receive interventions. The study was limited by the fact that only half the patient sample agreed to participate in the study and by the fact that participants in the study were not randomly assigned (the control group came from an earlier study). Moorey and Greer” reviewed the research literature that has evolved since the studies of Weisman et al. “‘18 They found the reported research sufficiently confounded by differences in type of psychotherapeutic intervention, statistical proce-

PSYCHOSOCIAL

RESEARCH

AND REHABILITATION

dure, and experimental design that it was not possible to determine if psychotherapywas helpful for the cancer patient. The question they originally asked was “Can psychotherapyimprove the (cancer) patient’s quality of life?” However, the very selection of this question illustrates one of the problems in the study of psychosocial aspectsof cancer; namely, that investigators select research objectives that are not necessarily consistent with the methodology being used. The authors seem aware of this becauseunable to answer their original question, they go on to state what should be the researchobjective for themselvesand presumably others: In future research, the question which should be asked is not whether psychotherapy is effective, but which specific, definable therapeutic procedures can be demonstrated to produce specified, measurable changes in psychosocial adjustment of patients with various kinds of cancer. (p 39).19

This more limited objective is also more consistent with the task of psychooncology of reducing or minimizing the psychiatric morbidity associated with cancer and its treatment. However, the simple assumption that reducing or minimizing psychiatric morbidity affects the quality of life of a person with cancer is an empirical question that needsto be verified, especially becausethe person remains a patient with cancer. To answer this question requires a study that involves determining if changes in depression, for example, impair a cancer patient’s functional status, the values they associate with their functional status, etc. However, most important is the fact that to have assumeda direct link between psychiatric status and quality of life reflects the mistaken assumption that mixing the various intellectual traditions that comprise the fields of study can be done without measurable consequences.In this case it confounded Moorey and Greer’s” evaluation of the contribution of psychotherapy to the welfare of the cancer patient. Social oncology has been quite an active areaof research, although not labeled as such. The work of Taylor et a12072’will be used to illustrate the field. Their work has focused on physician behavior as it occurs in the context of doctor-patient communication, accrual into clinical trials, and so on. Taylor et a120321 have found that physicians adopt one of two distinct philosophies, experimenter or therapist, in their approachto participat-

ing in clinical trials. Thesephilosophies have been found to differ in six ways including, the degreeto which the physician is committed to the welfare of the individual patient, as opposedto the group and future patients; whether the physician deals with the uncertainties inherent in clinical decision making by treating with known therapies, as opposed to withholding judgment and evaluating experimental therapies;the extent the physician is willing to share controversial medical information with a patient; the extent a physician acceptsclinical experience as a critical determinant of medical decision making; the extent the physician feels loyalty to current therapy, and the degree physicians see themselvesas individual practitioners or members of a group who share information about treatment options. Taylor’s20321work is important, not only because of the insight it has provided about an otherwise unexplored area of research, but also because of the methodological approach taken. Taylor, as a sociologist, is quite explicit about using sociological researchmethods in her work and has integrated her approach at every level of her activities. The work of Taylor et al*i also has potential practical applications. Because it provides insight into group dynamics, it could be used to rationalize the managementof cooperative clinical trial groups. Morrow’s** researchin the area of anticipatory nauseaand vomiting is another excellent example of a National Cancer Institute sponsored-research program that has succeededin the area of psychophysiological oncology. It has succeedednot only in terms of providing an effective intervention to deal with a disruptive behavioral consequenceof cancerchemotherapytreatment, but also becauseit deals with an issuethat is important to oncologists. The quality and relevance of the researchhas resulted in Morrow** leading one of the research basesAfor the Community Clinical Oncology Program.’ What Morrow** and others23324 have found is that standardbehavioral interventions (eg, progressive muscle relaxation procedures, systematic desensitization) can be used to reduce anticipatory and, in somecases,posttreatmentnauseaand vomiting. These procedureshave been found applicable for both children and adults under a number of different settings. Having established a reproducible and generalizable phenomena, Morrow and

196

IVAN BAROFSKY

his colleagues now have an opportunity to investigate the basic mechanismsunderlying the success of thesebehavioral interventions. Severalpotential mechanismshave been considered, including the role of nonspecific factors, counter-conditioning, distraction, perceived senseof control, and physiological relaxation. Black and Mort-0~~~suspect that all of these mechanismsmay be operative at somepoint and that individuals differ in the extent they use one or the other. Thus, future researchin the area will focus on studies of individual differences. To an oncologist, anticipatory nauseaand vomiting is a positive indicator for an intervention, whether behavioral or pharmacological, since not dealing with such behavior significantly increases the chancesof treatment noncompliance. What has been difficult for behavioral scientists and mental health clinicians to recognize is the importance of a positive indication for a referral decision. Too often behavioral factors are not shown to be sufficiently relevant to oncological practice to justify a referral (see below). One approachto this problem is to identify specific needs that a patient has, and to use these needs to make appropriate referrals. The work of Ganz et a126will be usedto illustrate this approach. The Cancer Rehabilitation Evaluation System (CARES) consists of 139 problems that cancerpatients may potentially have. Respondentsare asked to indicate the severity of any problem they have and also to indicate if they need help in dealing with the problem. The CARES evolved out of a competency-basedmodel of coping with cancer.27 Schagand Heinrich** define coping and the model as follows: Coping is defined as competent responsesto problematic situations. The model has three components, problem specification, responseenumeration, and responseevaluation. The first component identifies the domain of problems with which an individual must cope and provides a normative data base. The CARES was a fit step in the development of this model in that it would provide a method of data collection which could lead to a comprehensive data base of the problems with which cancer patients must cope. The instrumentmoves away from assessing cancer’s impact in terms of emotional distress and toward assessingmore specific components of behavior affected by cancer and its treatment. Our assumption is that a cancer-specific instrument will lead to better documentation of problems, better assessmentof individuals, and therefore, more appropriate and behaviorally-based interventions (p 6).

The last sentencein this quote is most important becauseit clearly states that the purpose of this need assessmentwas to develop a meansof justifying referral of cancer patients to mental health clinicians. This is done by determining what is normal coping for a specific study group (eg, breast cancer patients) and then to determine if a particular patient’s capacity to cope fits within the definition of what is normal. The definition used reflects a standardpsychometric approachthat uses a statistical criteria (eg, a scorehigher than 95% of the referencesample) to make a decision concerning the patient. The concern with this approach is that it requires that the clinician who made the referral to agreethat a coping scoreabove the 95th percentile is a sufficient reasonto refer a person to a mental health clinician. An alternative approach is reflected in the work of Houts et al.13,29They simply count the frequency of needsand allow the political process to determine if a referral should be made, resources allocated for services, and so on. The available studies suggesta consensusconcerningthe serviceneedsof cancerpatients.‘2*‘3,29-32 What seemsto be true is that a large proportion of any sample of cancer patients will report at least one unmet service need, that the proportion of patients who report such unmet service needs increasesas the patient becomesmore seriously ill, that the proportion who report psychosocial needs is modest (50% to 60%), and that the proportion who report a specific concrete need (eg, transportation) is low (approximately 15% to 25%). Although evaluating the importance of unmet needs has not been systematically studied, the measurement task is similar to a quality-of-life assessment-value clarification. Thus, there appearsto be at least two methods (statistical and social policy directed) for determining if a patient’s report of a needshould be considereda positive indicator for a clinical response. As Greer33argues, achieving humane treatment is a sufficient reason for involvement of psychosocial clinicians in the cancer patients’ care. However, most oncologists are not swayed by such an argument becausethey feel that the doctor-patient relationship they develop creates a humane and responsiveenvironment. In fact, they may seereferral to other clinicians as a disruption of this relationship. Rowland34lists (Table 3) a number of indicators of vulnerability to poor psychosocial ad-

PSYCHOSOCIAL

RESEARCH

Table 3. indicators of Vulnerability Psychosocial Adjustment Adjustment ACSIS

to Poor

Indicators

Medical

More physical symptoms More advanced cancer at diagnosis Doctor perceived as less helpful Dubious about effect of treatment Short time perspective about survival

Psychological/ psychiatric

Premorbid psychiatric history; suicidal ideation High anxiety Low ego strength More suppression More concerns of all kinds Helpless/hopeless attitude Feels victimized or blames others for cancer illness Alcohol abuse Poorer coping skills

Social

Data from

197

AND REHABILITATION

Lower socioeconomic status More marital problems More frequent background problems Expects or receives little support from others Rowland.34

justment that might be used to justify a referral. Unfortunately, virtually none of them (with the exception of suicidal ideation) constitutes a necessary and sufficient condition to justify a referral. Thus, not referring a patient for most of thesereasons would not make the oncologist feel that he or she was not doing his job. One example of researchwhich may affect the probability of a referral is the work of Spiegel et a135showing that group psychotherapy extends the survival of metastatic breastcancer patients. If this finding is supported by confirmatory studies, then referral for group psychotherapy would be consistent with the oncologist’s values of extending survival of the patient. Another area of research on positive indicators for psychosocial referral, which is only now beginning to be appreciated, is the role that a past history of physical and sexual abusehas on medical decision making.36 What is slowly becoming evident is that patients with this history are compromised in their ability to think through treatment alternatives. The doctor-patient relationship, with all of its psychological similarities to the parentchild relationship, is vulnerable to replicating the circumstances that has lead to the patient being abusedin the past. How such a history confounds both pat++ and clinician decision making is a le-

gitimate area of research with broad implications for clinical practice considering the high prevalence of an abusehistory.36 This review of psychosocial research in cancer rehabilitation, although selective, shows a rich, growing field of study that is increasingly becoming relevant to oncological practice. Although its onset was auspicious, psychosocialcancer rehabilitation researchis now poised for a major growth spurt. BASIC RESEARCH ISSUES IN PSYCHOSOCIAL REHABILITATION RESEARCH

Earlier it was mentioned that medical rehabilitation, by definition, is a restorative activity with efforts being extended to return a person with a deficit to some portion, if not all, of their former level of functioning. What is implicit in this definition is the assumptionthat restoration is possible. How such restoration will occur and if it can occur becomesthe fundamental researchquestion in rehabilitation. Thus, how is it possible to restore function in a chemotherapy-treatednervous system, surgerized muscle, or diseasedbone? These questions evoke issues in material sciences, biomechanics,transplant medicine, developmentalbiology, and pathophysiology. Part of the answer to this question seems straight forward; lost cells, muscle, and bone can not be replaced. Yet, demand by patients for rehabilitation reflects, in part, a belief that such lossescan be replaced, and the achievementsof modem medicine occasionally reinforce this view. Whether this demandreflects an unfulfillable wish or a lack of programmatic effort remains a legitimate question. To determine the limits of restoration in an areaof knowledge that is ever expending is a scientific issue (what is feasible), an administrative issue (how can scientific activities be directed to the problems implicit in this task), and a political issue (in terms of resource allocation). It is also likely that the functional status of a patient is maximized within the physical limits of the patient. The problem with this perspective is that it again sets the patient as the rate-limiting factor in rehabilitation, when this may not, in fact, be the case (see above). Still, how well a patient performs this role is the subject matter of psychosocial aspectsof cancer rehabilitation. Thus, it is again legitimate to ask how and to what extent

199

psychosocial cancer rehabilitation is possible? To answer thesequestions will involve developmental psychology, learning theory, group dynamics, and cognitive psychology, among other related fields. However, rarely do any of the sponsoringresearch agencies attempt to stimulate the development of the basic research that then can be applied in rehabilitation, in contrast to what has been done in areasof molecular medicine. Rather, the research sponsoring agencies see their task as applying already existing knowledge, a strategy that sometimes may be inappropriate considering what is known about a particular field. This was part of the difficulty that the Rehabilitation Program got into; application of knowledge that was inappropriate considering the complexity of the problems being studied. A classic example was the efforts to combat job discrimination against cancer patients (The Workable Project37). This national program consistedof a seriesof contracts that were terminated before substantial data were collected. The reasons for the termination were complex but included not understanding the group dynamics of a chronically ill person retuming to work. Unfortunately, very little has been learned since these efforts, so that if some sponsoring researchagency wanted to combat job discrimination in a rational and scientific manner, they would still have to go back and encouragethe basic research. Thus, as a result of such research miscarriages, the natural increments in knowledge characteristic of a researcharea is too often missing in the many areas of interest in psychosocial aspectsof cancer rehabilitation. Research that is available in psychosocial aspects of cancerrehabilitation has beenoverwhelmingly descriptive in nature, with very little study of mechanism. Yet, it is the understanding of mechanism that will permit the rational design of interventions. An example of the quality of the work required can be found in the area of doctor-patient communications.38 Detailed analysis of the exchangeprocesshas provided an empirical basis for the design of interventions. Another example is the work on patient preferences and decision making. 39-42This work has provided insight into the dynamics of patient decision making and preferences,an areaof increasing concern now that quality-of-life assessmentis playing an increasingly important role in clinical cancer medicine. A final

IVAN BAROFSKY

exampleis the work on how the attribution process affects perceptions of cancer.43-46Again, the investigators are linking a basic psychological processto a clinical observation. One of the impressive aspectsof psychosocial cancer rehabilitation is the large number of questions that can be asked in any area of research. Consider someof the issuesthat deal with the consequenceof cancer and its treatment on sexual functioning. For example, can a treatedcancer patient who has experienced disrupted sexual functioning be rehabilitated so that he or she can still have a satisfactorysexual life? What doesadequate sexual life mean?To what extent does the definition of satisfaction reflect the capacity of a person to cope with a more limited set of opportunities? What are the psychological processesinvolved in a person accepting limited functional and social opportunities? Are there physical limitations to the extent a person can psychosocially rehabilitate his or her sexual life? Thesequestionsraise issuesin recovery of neurological, physiological, or psychological functioning, as well as what constitutes “satisfaction or adequacy” of such functioning. Becausesatisfaction and adequacy are value-oriented issues they also are very much involved with quality-of-life assessment.Barofsky47has pointed out that there are basically four approachesto evaluating a quality-of-life outcome in a cancer treatment context. The first involves using a gold standard, which in most casesreflects a philosophical ideal (eg, the good life). In the caseof rehabilitation, this would mean restoring a person’s functioning to some ideal level of sexual functioning. A second approach would use population-basednorms. Thus, only if a person’s level of sexual functioning fit within 95% of the population would it be considered normal. A third approach uses all persons with the disease as the population, and again a statistical criteria is usedto define normal. Finally, there are treatment-related standards that can be used to determine a quality-of-life outcome. Here the objectives are to do no harm and to treat the person so that he or she can return to a pre-illness state. A comparisonof the functional change, from before to after treatment, would define a qualityof-life outcome. Thus, the very question of what is meant by “a satisfactory outcome” or being restored is a com-

PSYCHOSOCIAL

RESEARCH

AND REHABILITATION

199

plex one that can be answeredin at least four different ways. Becausethere are many problem areasin psychosocial cancerrehabilitation, it is easy to seehow complex a task it is to sponsorresearch in this area. What is also evident is that a political consensusneeds to develop about what is a reasonable outcome of cancer rehabilitation, in general, and psychosocial cancer rehabilitation in particular. The absenceof this consensusis part of the reasonthat sponsoredresearchin this arearemains at a minimum. SUMMARY

This article has tried to describe the current status of psychosocial researchin the rehabilitation of the cancer patient. It attemptedto weave together the author’s perspective of how decisions early in the history of sponsored research programs, particularly by the National Cancer Institute, combined with a limited knowledge baseled to limited growth of the Rehabilitation Program. Thus, the current statusof psychosocial cancerrehabilitation can be reasonably attributed to the decision at the National Cancer Institute to encouragethe development of cancer rehabilitation services, rather than to aggressively expand the knowledge base. Although, in retrospect, it is possible to criticize these decisions, in fact, they were legitimate choices among a wide range of options. It was also not possible in this article to discuss many topics in psychosocial cancer rehabilitation,34 particularly differences in psychosocial adjustment as a function of type of cancer. What the study attemptedto do was to confront the impression that cancerrehabilitation, in general, and psychosocial cancer rehabilitation, in particular, are ancillary activities that can receive a secondary level of resource allocation and support. The point was made that determining if psychosocial rehabilitation is possible raises as many basic research

questions as does understanding how chemotherapy works or how a malignancy develops. All three areasask legitimate basic researchquestions on how the body works and how behavior changes. Today, 10 years after the start of the original Rehabilitation Program, psychosocial cancer rehabilitation is an establishedfield of study and an integral part of most major oncology services. Now, as stated,what is neededis an expansionof its knowledge base. Someof the issuesthat are deserving of support include studies on cosmesis,to what extent voluntary processescan compensateloss in speechand swallowing functions, functional evaluation following alternative surgical procedures,47what is the quality of life following long-term survival of cancer, what are the group dynamics following return of a cancer patient to a work site, and so on. Each of these researchquestions can be guided by the same model developed for cancer control research by Greenwald and Cullen. Other more general approachesinclude recognizing and promoting the preventative dimension of cancer rehabilitation49 and developing the art and practice of psychometric assessmentof psychosocial aspects of cancer rehabilitation, just to name two such areas. The American Cancer Society also recently sponsoreda meeting to review and set new directions for psychosocial rehabilitation research.“’ What may also be evident from this review is the critical interdependenceof science progress and scienceadministration. Paradoxically, much lessis known about how to successfully administer a science program then to do science. If the early history of the Rehabilitation Program has any longterm consequenceit should be by promoting the study of science administration, in and of itself. ACKNOWLEDGMENT Dr. Carolyn Gotay’s helpful comments are acknowledged.

REFERENCES 1. Gladwell M: A spectacular failure: Cure is illusive, but effort built biotech industry. Wash Post Health, December 3, 1991, pp 5 2. Gotay CC, Yates JW: Emerging trends in cancer rehabilitation research. Proceedings of the 5th National Conference on Human Values and Cancer-1987. Atlanta, GA, American Cancer Society, 1987, pp 18-22 3. Harvey RF, Jellinek HM, Habeck RV: Cancer rehabili-

tation: An analysis of 36 program approaches. JAMA 247: 2127-2131, 1982 4. The National Cancer Act of 1971, Public Law #92-218. Section 409 5. National Cancer Rehabilitation Planning Conference Chairman’s Report: National Cancer Program, US Department of Health, Education, and Welfare, National Institutes of Health, National Cancer Institute, June, 1973

IVAN BAROFSKY 6. Community Mental Health Centers Act, Amendments, Public Law #95-622, Section 403 7. Health ResearchExtension Act of 1985,Public Law #99158, Section 412 8. Gunn AE: Cancer rehabilitation: An overview, in Gunn AE (ed): Cancer Rehabilitation. New York, NY, Raven, 1984, p 3 9. Community Clinical Oncology ProgramInvolved in a Variety of Cancer Control Protocols: The Clinical Cancer Letter 14 (12):Bl, B3-B5, 1991 10. Report of the Task Force on Medical Rehabilitation and Research, Hunt Valley, MD, June 28-29, 1990, Hunt Valley Inn 11. FeldmanJJ: Workability of the aged under conditions of improved mortality. Milbank Q 61:430-444, 1983 12. Mor V, Masterson-Allen S, Houts P, et al: The changing needs of patients with cancer at home: A longitudinal view. Cancer 69829-838, 1992 13. Houts P, Yasko J, Kahn SP, et al: Unmet psychological, social, and economic needsof personswith cancer in Pennsylvania. Cancer 58:2355-2361, 1986 14. Holland JC, Rowland JH: Handbook of Psychooncology: PsychologicalCam of the PatientWith Cancer. New York, NY, Oxford, 1989 15. Cassileth BR, Lusk EJ, Miller DS, et al: Psychosocial correlates of survival in advancedmalignant disease?N Engl J Med 312:1551-1555, 1985 16. Cassileth BR, Walsh WP, Lusk EJ: Psychosocialcorrelates of cancer survival: A subsequentreport 3-8 years after cancer diagnosis. J Clin Oncol 6:1753-1759, 1988 17. Weisman AD, Worden JW: Coping and vulnerability in cancer patients. ResearchReport. Project Omega, Department of Psychiatry, Harvard Medical School, MassachusettsGeneral Hospital, Boston, MA, 1977 18. Weisman AD, Worden JW, Sobel HJ: Psychosocial screening and intervention with cancer patients: ResearchReport. Project Omega, Department of Psychiatry, Harvard Medical School, MassachusettsGeneral Hospital, Boston, MA, 1980 19. Moorey S, Greer S: Psychological therapy for patients with cancer:A new approach. Washington,DC, American Psychiatric, 1989 20. Taylor KM: The doctor’s dilemma: Physician participation in randomized clinical trials. Cancer Treat Rep 69:10951097, 1985 21. Taylor KM, Shapiro M, Skinner AJ: Some thoughts on the future of clinical trial groups in cancer. Cancer Treat Rep 71:434-435, 1987 22. Morrow GR: Chemotherapy-relatednauseaand vomiting: Etiology and management.CA 39:89-104, 1989 23. Burish TG, Carey MP, Krozely MG, et al: Conditioned side effects induced by cancer chemotherapy: Prevention through behavioral treatment. J Consul Clin Psycho155:42-48, 1987 24. Redd WH, JacobsenPD, Die-Trill M, et al: Cognitive/ attentional distraction in the control of conditioned nauseain pediatric cancerpatients receiving chemotherapy.J Consul Clin Psycho155391-395, 1987 25. Black PM, Morrow GR: Anticipatory nauseaand emesis: Behavioral interventions, in WatsonM (ed): CancerPatient

Care: PsychosocialTreatment Methods. Cambridge, England, Cambridge University Press, 1991 26. Schag CAC, Heinrich RL, Aadaland RL, et al: Assessing problems of cancerpatients: Psychometricproperties of the cancer inventory of problem situations. Health Psycho19:83102, 1990 27. Meyerowitz BE, Heinrich RL, Schag CAC: A competence-basedapproach to coping with cancer, in Burish TG, Bradley L (eds): Coping With Chronic Illness: Researchand Applications. New York, NY, Academic, 1983 28. Schag CAC, Heinrich RL: Cancer Rehabilitation Evaluation System, Los Angeles, CA, CARES Manual, 1988 29. Houts PS, Yasko JM, Harvey HA, et al: Unmet needsof personswith cancer in Pennsylvania during the period of terminal care. Cancer 62:627-634, 1988 30. Lehmann J, DeLisa J, Warren D, et al: Cancer rehabilitation: Assessmentof need, development, and evaluation of a model of care. Arch Phys Med Rehabil 59:410-419, 1978 31. Moore V, Guadagnoli E, Wool M: An examination of the concreteservice needsof advancedcancer patients. J Psychosoc Oncol 5:1-17, 1987 32. Siegel K, Raveis VH, MesagnoFP, et al: The needsand service requirementsof seriously ill patients in community settings. Presentedat the annual meeting of the American Public Health Association, Boston, MA, November 13-17, 1988 33. Greer S: Forward, in Watson M: Cancer Patient Care: PsychosocialTreatment Methods. Cambridge, England, Cambridge University Press, 1991, p 1 34. Rowland JH: Psychosocialissues in the cancer patient: Impact on care, in McGarvey CL (ed): Physical Therapy for the Cancer Patient. New York, NY, Churchill Livingstone, 1990, p 142 35. Spiegel D, Bloom JR, Kraemer HC, et al: Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet 14:888-891, 1989 36. DrossmanDA, LessermanJ, Nachman G, et al: Sexual and physical abusein women with functional organic gastrointestinal disorders. Ann Intern Med 113:828-833, 1990 37. Barofsky I: Work: Issues and concepts, in Barofsky I (ed): Work and Illness: The Cancer Patient. New York, NY, Praeger, 1989, p 16 38. Roter DL: Patient participation in the patient-provider interaction: Effects of patient asking on the quality of interaction, satisfaction,and compliance. Health Ed Mono 5:281-315, 1977 39. Llewellyn-Thomas HA, SutherlandHJ, Tibshirana R, et al: Describing health states:Methodological issuesin obtaining values for health states. Med Care 22:543-552, 1984 40. McNeil BJ, PaukerSG, Sox HC, et al: On the elicitation of preferencesfor alternative therapies. N Engl J Med 306: 1259-1262,1982 41. O’Connor AM: Effects of framing and level of probability on patients’ preferencesfor cancer chemotherapy.J Clin Epidemiol 42:119-126, 1989 42. Siminoff LA, Fetting JH, Abeloff MD: Doctor-patient communication about breast cancer adjuvant therapy. J Clin Oncol 7:1192-1200, 1989 43. Taylor SE, Lichtman RR, Wood JV: Attributions, be-

PSYCHOSOCIAL

RESEARCH

AND REHABILITATION

liefs about control, and adjustmentto breast cancer. J Pers Sot Psycho146:489-502, 1984 44. Gotay CC: Why me? Attributions and adjustment by cancer patients and their mates at two stages in the disease process. Sot Sci Med 20:825-831, 1985 45. Easterling DV, Leventhal H: Contribution of concrete cognition to emotion: Neutral symptoms as elicitors of worry about cancer. J Appl Psycho174:787-796, 1989 46. Curbow B, Andrews RM, Burke TA: Perceptionsof the cancerpatient: Casual explanations and personal attributions. J PsychosocOncol 4:115-134, 1986

201

47. Barofsky I: Quality of life assessmentin musculoskeletal surgery. Theor Surg 7:27-30, 1992 48. Greenwald P, Cullen JW: The new emphasis in cancer control. J Nat1Cancer Inst 74:543-551, 1985 49. Gotay CC, Yates J: Meeting summary: Workshop on preventative rehabilitation researchto reducecancermorbidity. National Cancer Institute, Division of Cancer Prevention and Control, September 11, 1987 50. Burish TG: Behavioral and psychosocial cancer research:Building on the past, preparing for the future. Cancer 67:865-867, 1991