ISSUES IN THE UTILIZATION AND EVALUATION OF SOCIAL WORK TREATMENT

Abstract

Five issues regarding utilization and evaluation of social work treatment are considered with the goal of understanding their impact on development of sound treatment technologies : (1) Outcome vs. Process Orientation, (2) Demonstrated Effect vs. Need Fulfillment, (3) Cost Effectiveness vs. Individual Commitment, (4) Individual Constructs vs. Social Context, and (5) Practice Art vs. Science. It is assumed that contributions to the resolution of these issues require specification of desired outcomes expected to result from specific professional activities.

“Some values are…like sugar on the doughnut, legitimate, desirable, but insufficient, apart from the doughnut itself. We need substance as well as frosting.” — Ralph Tyler Fleweling.

In any planning endeavor involving a designated social problem it is necessary to ask at least two basic questions : (1) What techniques are available for dealing with the problem? and (2) How effective are these technologies in producing a satisfactory outcome?

While these basic questions are important, other factors such as the political, financial, and ethical feasibility of a problem-solving technology are influential in determining the final outcome of a planning endeavor. It is within this framework of cost, politics, ethics, and the availability of other competing technologies that the profession of social work faces the test of developing effective problem solving technologies.

The primary investment of the social work profession has been as providers of direct services to various groups with several types of problems. The direct services include social casework, group work, and individual and group psychotherapy. The actual content of these direct-service activities has not been well defined, though it is generally acknowledged that they are reliant on the use of relationship, interpersonal influence and skilled interviewing techniques to bring about client changes.

In the past few years there has been a good deal of controversy in social work education regarding the emphasis that should be placed on training for direct vs. indirect services. Since the mid-sixties direct services have been under attack. The Community Service Society of New York, one of the oldest, largest, and most respected social work agencies in the United states had chosen to reorient their programs in this area — a decision which has recently been modified under the leadership of Alvin Schorr.

Underlying much of the ferment and discontent that has been expressed with the direct services are five issues in the utilization and evaluation of this intervention mode. The purpose of this paper is to explicate these issues and the choices they pose for the direct services. Generally these issues may be described as “outcome vs. process.” The focus on outcome and process leads to four other major issues. (See Fig. 1.)

Figure 1.

An outcome orientation gives rise to two major types of issues. First, “demonstrated effect” vs. “need fulfillment”; and second, “cost effectiveness” vs. “individual commitment.” A process orientation generates two other major issues : “individual constructs” vs. “social context”; and “practice art” vs. science.

These issues will be discussed from a technological point of view, toward the end of specifying how differing types of treatment intervention lead to specifiable solutions to specifiable problems.

1. Outcome vs. Process Orientation

The first and most general issue, outcome vs. process orientation, deals primarily with two competing orientations toward the purpose of social work interventions. An outcome orientation demands that the goals of intervention be specified before the intervention is undertaken; a process orientation does not specify these goals but instead asserts that what takes place in the course of treatment is in fact the purpose of the endeavor. Kasius states it thus:

Problems are seen to result from a destructive use of relationships. It follows, therefore, that the central factor in casework help is a relationship — an experience in form — in which the client can experience a new and constructive way of using himself and the other person (Kasius, 1951, p. 11).

This issue involves the great difficulty of specifying the desired outcome of a human interaction problem-solving process. The complexity of doing this is typified in questions that immediately confront the direct service worker when he sees a client such as the following : should this client be able to function more effectively as a result of treatment; by what criteria do I assess effective functioning; for how long should he function better; during treatment or after treatment; what if he functions more effectively but is unhappier; if all the clients of my marriage counselor colleague eventually get divorced is he producing a desired outcome?

As in most complex endeavors, even if one adopts an outcome orientation in social work treatment the risk of goal displacement is high. It is tempting to focus on the abstract nature of the treatment process independent of goals and often at the expense of efforts directed to their specification.

Schwartz and Sample, in their article “First Findings from Midway,” note:

We rejected the frequently advanced notion of defining the social worker’s job on the basis of the kind of client, difficulty of problem, or emergent nature of the situation as a basis for identifying the essentially professional component. We adopted instead the idea of professional function which we defined in relation to individual cases to include : 1) study; (2) diagnosis; (3) case classification; (4) the laying out of the plan of treatment; (5) administration of the treatment plan through supervision and direction of workers; (6) provision of direct treatment when considered appropriate because of the need for establishing a treatment relationship; (7) change of diagnosis, case classification, or treatment plan as appropriate (1967, p. 119).

This description of professional function is exceedingly ambitious when one considers the diversity and number of problems with which a worker must deal. It should be noted that the establishment of appropriate goals as a mandate of professional function is perhaps only implied in the task of laying out the treatment plan. The Midway Project represents a major step in the direction of a sound approach toward the development of process research. Yet one must necessarily question the validity of a process which is non-goal directed.

Among the major current criticisms of the profession, failure to specify goals is of particular interest in view of the initial high dropout rate evidenced in social work treatment (Ripple, 1957) and the often-stated lack of congruence between goals of the client and goals of the professional. This lack of agreement on goals is illustrated in experience with individuals who drop out of social work treatment against their therapist’s recommendation after having made an initial commitment to a minimum of four therapeutic sessions. In a recent look of this phenomenon, Jones (1972) found that case records of dropouts contained little information about the patient contact while the records of individuals who continued in treatment contained detailed reports of the therapeutic activities that occurred during the treatment period. It may be hypothesized that the therapists and the dropouts were never able to establish a common ground for being there. Though it is obvious that individuals often enter a therapeutic relationship due to outside pressures, it would seem that these pressures did not come to light in the dropout group.

Given this type of critical comment, social workers practicing between 1930 and 1950 have often noted that process is the goal of treatment. The writings of Jessie Taft (1933) and Cora Kasius (1951) are sometimes like Alice’s conversation with the Cheshire Cat:

“Would you tell me, please, which way I ought to go from here?”

“That depends a good deal on where you want to get to,” said the Cat.

“I don’t much care where —” said Alice. “Then it doesn’t matter which way you go,” said the Cat.

“— so long as I get somewhere,” Alice added as an explanation.

“Oh, you’re sure to do that,” said the Cat, “if you only walk long enough.” (Carroll, 1916, p. 60)

In more recent years the importance of specifying goals in the treatment endeavor has been recognized. Yet the past emphasis on process as a goal still exerts strong influence in the practice literature. Thus, in attempting to combine the emphases on process and goals in casework Perlman (1957) notes somewhat ambiguously that:

The end of this process is contained in its means:

to so influence the client-person that he develops effectiveness in coping with his problem and/or to so influence the problem as to resolve or vitiate its effects. (p. 5)

Perlman’s comment contains the crux of the issue of an outcome vs. a process orientation. She implies that the process is the goal. However, if that is the case, then it is extremely difficult if not impossible to measure success or to understand failure to achieve it. Moreover, the question of what specific process is most appropriate is irrelevant (e.g.,Freudian analysis, Transactional analysis, Primal Therapy, Social Casework, simple problem solving, or perhaps a process that is important solely for its ceremonial (Frank, 1963) or supportive (Halmos, 1967) value.

The focus on process alone leads to an obscuring of possible outcomes, and it encourages use of a treatment interaction that may produce undesirable consequences unbeknownst to, and in direct conflict with, the ideology of the service giver.

How frequently does the juvenile offender enter into a therapeutic relationship simply to stay out of the youth house? Of course, staying out of the youth house might be a laudable goal. I simply want to emphasize that failure to specify goals and to relate process to goals may lead to bootlegging illegitimate goals under the auspices of social work process.

2. Demonstrated Effect vs. Need Fulfillment

While the process vs. outcome dilemma is characteristic of much of direct practice there are some practitioners who employ an outcome orientation in their work. The adoption of an outcome orientation, however, gives rise to other problems as represented in the issue of demonstrated effect vs. need fulfillment. The essential problem here is not in selecting objectives but rather in selecting criteria against which “success” is measured. The issue of outcome vs. process orientation refers primarily to the organization and direction of one’s work. In contrast, the issue of demonstrated effect vs. need fulfillment refers to the justification for the continued effort.

Need fulfillment involves the achievement of a specific outcome. Demonstrated effect is also concerned with the achievement of a desired outcome in treatment, but in addition, it seeks to determine the extent to which an outcome may be attributed to the influence of the social work effort. Based on a review of 19 studies of social work treatment with psychologically-based problems, Segal (1972) reported a mean improvement rate of 72.3 %. Given previous reports by Eysenck (1952) and Levitt (1957) indicating “spontaneous remission” rates of 72 % and 72.5%, respectively, for psychologically-based problems in similar population groups, Segal (1972) observed that the average improvement rate may, in fact, reflect a similar phenomenon.. However, recent research (Bergin, 1971; Subotnik, 1972) has raised question about these observations. Bergin (1971), in reviewing the reported “spontaneous recovery” rates for 14 control groups seeking help from non-social work practioners for psychologically-based problems, found a mean 25 % improvement, a finding much lower than Eysenck’s (1952) and Levitt’s (1957). This suggests that Segal’s (1972) finding of a mean improvement rate of 72.8 % for social work treatment of psychologically-based problems may not in fact be attributed solely to spontaneous remission but either to the effectiveness of social work intervention or to a rating process that is consistently biased (Garfield, Prager, and Bergin, 1971; Segal, 1972).

It should be noted, however, that a sound interpretation of scientific findings is contingent upon well designed and replicated studies. Given these two criteria, the observation should be made that the data on the outcome of social work treatment (assuming the validity of the rating process, an assumption that is open to question) is consistent in at least one area — treatment of psychologically-based problems. Further review (Segal and Black, 1975) of the outcomes of 25 additional studies (n = 10,264) of treatment offered by social workers to people with psychologically-based problems yielded a weighted mean improvement rate of 67 %¹. Comparing these findings with Bergin’s (1971) finding of 25% spontaneous remission allows us to view social work treatment for psychologically-based problems in a more positive vein. However, these findings on “demonstrated effects” cannot be conclusive. Well-controlled studies are needed along with more rigorous research on the development of reliable and valid indices of improvement and measurements of the effects of different findings.

Demonstrated effects in other areas of social intervention are even more equivocal. Unfortunately, Fisher (1973) has chosen to condemn the total casework process on the basis of a limited review of studies in these other areas. As Segal (1972) has pointed out, since the 1930’s, well-designed studies have indicated the failure of social work attempts to cope with delinquency prevention (Berleman and Steiner, 1967; Berleman, et al., 1972; Meyer, Borgatto and Jones, 1965; Powers and Witner, 1951). While one recent well-designed project (Baron, Feeney, and Thornton, 1973) has demonstrated a partial success at delinquency prevention, the findings of the study must be carefully considered in view of the large number of other reports of failure in this area. Studies on the effectiveness of social work intervention in problems of delinquency and in programmes for the aged and public assistance populations also continue to report mixed and unclear findings (Segal, 1972; Segal and Black, 1975).

Finally, Beck and Jones (1973) found that when agencies use traditional psychologically-based methods (e.g. Family Service Agencies) for the purpose of ameliorating social problems, there tends to be less improvement in an overall outcome rating based on average change scores. This type of variance in the overall improvement rating is an indicator of its validity as an index of treatment outcome especially in view of the more equivocal or negative results of social treatment efforts with social problem populations.

Measurement Problems and Demonstrated Effect

The concept of demonstrated effect has been bandied about by social work researchers who have frequently castigated their practitioner colleagues for their inability to produce such effects (Fisher, 1973). The researchers, however, have been equally inept in their efforts to design adequate devices to measure these effects.

In dealing with social problem populations the solution to the measurement of effects appears to be rather simple. For example, success for the delinquent or probationer is his not getting arrested, for the AFDC father success is leaving the welfare rolls for gainful employment, for the pre-delinquent success is graduation from high school. The problem with measuring effect by these criteria is that the specified outcome cannot be completely summarized by a present vs. an absent condition because the outcome is itself a multi-faceted characteristic of an individual’s psychology, social life, or academic and vocational ability.

When the desired outcome is a multi-faceted characteristic then the question arises as to how much change is sufficient. Failure to specify the degree of change that is desired in a group of subjects may lead to this determination being made by other factors which often influence selection of a given sample².

The larger one makes his study sample, the more likely he will be to find a statistically significant difference between treatment and control groups, since larger samples detect smaller differences between groups (whether or not such a difference is substantively significant). Conversely, the difficulty and expense of obtaining large samples for experimentally designed outcome studies means that research must demonstrate that there are quite large differences between groups in order to show significant effects.

Several authors have dealt with the important question of reliability of rating in social work research. That is, they have had several individuals rate the same outcome material with a high degree of agreement between them (e.g., see Kogan, Hunt, and Bartelme, 1953; and Reid and Shyne, 1969). They have also determined whether current outcome assessments can be used to predict future events of similar character (Kogan, Hunt, and Bartelme, 1953) i.e., whether the measurement instrument has any predictive validity. However, no efforts have been made to determine whether the items included in social work outcome assessment are valid, that they actually share a component that can be called “social functioning,” or some other theoretical construct.

This lack of effort in construct validation is a serious problem which may account for the recent failure of a research group to comprehend the result obtained in their study of social work input into the encouragement of gifted children living in deprived environments. In McCabe, et al.’s (1967) study of intellectually superior children in a socially deprived area it was hypothesized that such children “do not tend to achieve their potential or maintain their relatively high level of intellectual and academic performance because of the detrimental effects on their ego structure of the pathogenic environment to which they are exposed (p. 232).” The focus of the project was to prevent with the aid of social work intervention techniques the expected further impairment of the “ego functioning” of this group. Separate analyses of the study results were completed for the total sample and for the Black and Puerto Rican groups within the total sample. The results indicated no difference in the overall functioning of the experimental and control cases — i.e., 67% of the experimental and 72% of the controls improved in their overall “ego function” rating. However, a differential pattern in the Black and Puerto Rican subsamples was observed that tended to account for the lack of observed difference between the experimental and control assessments for the total group. While the Puerto Rican group tended to follow the predicted pattern hypothesized in the study (i.e. 69% of the Puerto Rican experimental cases improved their functioning compared to only 14% of the control cases) the Black subgroup results were unanticipated (i.e., 64% of the experimental group and 93 % of the control group improved). These variations are explained somewhat by Dohrenwend and Dohrenwend’s (1969) study of levels of psychopathology manifested in different ethnic groups. Using the 22 Langner items found to be most predictive of severe psychological disturbance in the Midtown Manhattan Study, they compared the responses of Black, Puerto Ricans, Jewish and Irish subjects on this rating scale. To their dismay they found that while the Puerto Ricans scored high on the psychopathology scales (indicating a more severe level of psychological disturbance) the other groups did not. They were particularly puzzled about the failure of the Black group to produce higher symptom scores than the Jewish and Irish groups since they believed that the former group experienced a greater amount of social stress than the latter two groups. Exploring their data in greater detail, the Dohrenwends analyzed the extent to which the four ethnic groups in their study viewed the symptoms inquired about in the 22 Langner items as “socially undesirable.” The results of their inquiry indicated that Puerto Rican respondents tended to view the symptoms reported as less socially undesirable than the other three groups studied.

Returning to the McCabe study, it should be noted that a majority of the outcome measures used attempted to assess changes in the “ego functioning” of the study group. However, all the “ego functioning” measures were probably more related to a construct better named “propensity to respond in a socially desirable manner” than to the “ego functioning” construct. This observation offers the best interpretation of McCabe’s results in view of the Dohrenwends’ findings. The test of reading performance used by McCabe showed that both the Black and Puerto Rican experimental sub-groups improved significantly more than their respective controls. Other academically-oriented tests showed no significant differences from their controls in either sub-groups.

At first glance, it may appear that social work interventions actually improved the ego functioning of the Puerto Ricans, because their responses are less distorted by a tendency to give socially desirable responses. In fact, though, it may have been that the Puerto Rican experimental group learned how to respond in a socially desirable manner through participating in the programme. However, the impact of the programme in enhancing the reading skills of the total experimental group must not be overlooked in a society in which reading ability is basic to the fulfilment of must achievement goals. Reading skill, unlike “ego functioning,” can be measured clearly and is less distorted by social desirability interactions that may enter the measurement procedures.

Building a sound professional technology requires the ability to demonstrate the effect of professional interventions. Yet, the difficulty that researchers and clinicians have experienced in achieving this goal appears to have created a defensive spirit between the two groups with each faction attempting to cover its inadequacies rather than co-operating to achieve satisfactory solutions.

Need Fulfilment : Continued Service in the Absence of a Clearly Demonstrated Effect

From a policy perspective, it might be expected that evaluative research would have an impact on the continued funding of intervention programmes. However, because of the thorny problems of assessment in evaluative research, the absence of a clearly demonstrated effect is not sufficient justification for discontinuing a programme effort.

The rationale for continued funding of programmes for which research has consistently produced negative findings appears questionable. For example, since the late 1930’s, research of relatively high quality in the area of delinquency prevention has indicated the absence of any positive effects of social work intervention except for one project which had mixed results (Baron, Feeney, and Thornton, 1973). Why, then do traditional efforts at delinquency prevention start up each year?

The continuance of social work programmes, whether of proven effectiveness or not, may be attributed to need fulfillement. That is, these programmes fulfill societal, and client needs whether or not there are identifiable outcomes that can be related to articulated goals. Some people are required to use services in order to continue to receive other benefits from agencies. For example, many welfare clients have traditionally felt the obligation to accept services in order to continue to their cash grants. Moreover, many consumers continue to use social service programmes because they find they are helping them to deal with everyday problems in living despite the fact that we do not understand exactly how or why this occurs.

The legitimation of services through licensure of professionals creates a presumption of confidence in the service provider’s ability to affect an individual’s problem situation. In fact, belief in the skill of the service provider is a powerful therapeutic device (Frank, 1958) and is perhaps a reasonable justification; along with the need for public protection for continuing to license therapists who deal with these problems.

However, consumer demand only partially explains the continued use of social services. It is attributable also to societal need fulfillment and professional need fulfillment. Societal need fulfillment has two major aspects: social control and organizational survival. One of the primary social mandates of the mental health professions is the restoration of equilibrium in the population and the prevention of social conflict. The street worker in a delinquency prevention programme often finds his role to be that of a “safety valve.” For example, programme workers with delinquent gangs continuously absorb hostility that might otherwise be directed toward the community.

Organizational survival can also be considered under the rubric of societal need fulfillment. Any social organization tends to prepetuate itself and further its own growth. Program people constitute on organized political force with an altruistic goal of offering services. On the other hand the consumers of mental health services are a rather disorganized group. It is not infrequently claimed by agencies that they are successful because of the achievement of community acceptance. It should be noted, however, that such community acceptance may be as much a result of the actions of a skilled executive as of the effectiveness of the service program.

The professional’s personal needs are also part of the reason for continued offering of service in the obsence of a clearly demonstrated effect. Three particular professional needs might be stated in this light: (1) the continued use of their professional skill; (2) the altruistic need to help others, and (3) the need to engage in an intellectually stimulating endeavor.

The need for continued use of their skills is a problem faced by most professions. Once individuals have trained in the professions and develop a given set of skills, they find it necessary to use them for their economic survival.

The second major personal need of the social work professional is expressed in the altruistic ideology of the field: professionals must do whatever they can regardless of current knowledge regarding· the effectiveness of their intervention strategies. The assumption underlying this activity has been that whatever we do it will, at best, be helpful and, at worst, of no consequence. However, given the assumed power of our techniques to deal with the most difficult social and psychological problems, it seems odd that we should have not considered the possibility of negative consequences resulting from our intervention strategies. Cartwright (1956) in reviewing the Barron and Leary (1955) evaluation of psychotherapy noted that while no differences were observed between experimental and control groups, the results indicated a greater diversity of behavior within the experimental compared to the control group. This observation would indicate that while the average person in both the treatment and control groups improved about equally, some individuals in the treatment group benefited from the experience and others may have been harmed by it. (Similar findings have been reported by Bergin, 1971). From a technological perspective Cartwright (1956) tended to attribute these results to differences in therapist experience.

From the perspective of evaluating social work one should not go over-board in condemnation of intervention efforts on the basis of reports of what appears to be negative effects of treatment. It must be remembered that science is built on replicated findings from many studies. For example, Fisher’s (1973) observation of work done by Bleakner, Bloom, and Nielsen (1971) is a rather careless analysis which leaps to conclusions too quickly. Fisher notes that in the Bleakner, et al study a group of aged persons who received protective service because they had difficulty caring for themselves manifested a higher death rate than a randomly matched control group. While this is in fact what Bleakner, et al found, contrary to Fisher’s report the death rate difference between the treatment and control group was not statistically significant. A test of the differences between the experimental and control groups in the Bleakner, et al. sample yields a X²=2.37, p 30. That is, these results could have been obtained in approximately one out of three samples of similar size by chance alone.

Social work services may indeed have a negative effect on their clientele. However, such negative effects must be carefully validated with the goal of developing adequate treatment technologies rather than used to make an argument for the total ineptness of the field. “Next to truth, a confirmed error does well (Ben Johnson : Bartholomew Fair, 1614).”

Another factor that influences the social work professional’s choices in treatment methodology is his need of opportunities for “self-fulfillment” in the therapeutic endeavor. “Self-fulfillment” seems to be tied to the intellectual manipulation of varying conceptions of psychotherapy. Such “self-fulfillment” is often absent in work with the unmotivated client or the psychotic patient and may account for the relatively low percentage of trained workers involved with these types of clients.

The popular book, I Never Promised You a Rose Garden, illustrates the different types of personal need-fulfillment required by the social work professional (Green, 1964). The book may to some extent be considered a statement of ideology. In a Durkheimian sense it maintains and supports the norms associated with psychotherapeutic treatment between patient and therapist. Cure takes place within the intellectual confines of an organized ideology. With this ideology the therapist fulfills her altruistic need to help the patients and, most importantly, the patients needs are met. Treatment in this frame of reference seems to involve the acceptance of the ideology as a means of coping with life or as a cure.

What is important in considering the issue of demonstrated effect and need fulfillment is the question of whether the client’s needs predominate and are met.

3. Cost Effectiveness vs. Commitment to Individuals

The crux of this issue is whether outcome measures and professional efforts will be focused solely on how an individual improves in treatment or will emphasize the value of changes in the larger community group.

Social work’s commitment to the unique worth of the individual has to some extent placed it at a disadvantage in offering services to the larger community group. The emphasis in social work has been on providing individual treatment.

There would be little argument in the field with the policy of keeping an individual in treatment as long as necessary to achieve a satisfactory solution to his problem. This approach contrasts sharply with the public health model whose outcome referents — primary, secondary and tertiary prevention — refer to changes occurring in the larger community group. In all three types of preventive efforts outcome is expressed in terms of observed changes in specified rates. In the case of primary and secondary prevention the changes are directly related to changes in the incidence and prevalence rate, respectively, of any behavior syndrome that has a specifiable onset, duration, and termination. In the case of tertiary prevention changes are viewed as modifications in the rate of occurrence of disability associated with a given phenomenon. More specifically, primary prevention refers to reducing the proportion of new occurrences of the phenomenon of interest in a population. Secondary prevention refers to reducing the total proportion of the population evidencing this phenomenon and is usuallly operationalized in terms of shortening the duration of a given disorder. Tertiory prevention usually refers to efforts to reduce the occurrence of chronic disability often due to iatrogenic effects associated with the phenomenon of interest.

Two principal characteristics associated with the public health model are “efficiency” and “risk,” both of which result from a group focus as opposed to an individual focus. The combat surgeon’s job is to keep his army’s death rate as low as possible. In accomplishing his task he makes definite judgments about the use of his time and efforts. He chooses his cases from among the wounded on the basis of their current risk of death and also on the basis of whether or not he can possibly offer help. That is, he must decide whether a prospective case is too far gone to command his attention given the pressing need of others in the group.

The resources of social work are limited and therefore careful decisions must be made about the investment of professional effort.

This argument should not lead to the conclusion that all is rosy with the preventive model. Prevention implies knowledge of necessary conditions for the occurrence, continuation and exacerbation of a given phenomenon. Our knowledge in this area is still at the rudimentary level. In the mental health field, however, several lines of research are currently in progress that point to possible ways of increasing the cost effectiveness of social work efforts.

Primary prevention efforts at generalized mental health education seem highly questionable and destined to failure (Cumming and Cumming, 1958). On the other hand, some work has been done that enables us to begin to designate high-risk target groups. For example, the work of Brown and Birley (1968), Dohrenwend (1973), Meyers, Linderthol and Pepper (1971), and Rahe (1971), indicated that the occurrence of an increasing number of life events such as the death of a spouse, a wedding, a daughter’s graduation, and a business merger or failure are associated with the onset of illness (both physical and mental). Perhaps we may also begin on the basis of these research findings to advocate postponements of certain events in order to modify risks.

Anthony (1972) has recently set up a risk-crisis monitoring system in schools where children are assigned risk scores according to gross rating of the genetic, reproductive, constitutional, developmental, physical health, environmental, and traumatic factors assumed to relate to the occurrence of mental disorder. These children are then monitored with greater or lesser frequency according to their risk score. This system of course makes two major assumptions: (1) that we have reliable indicators of risk; and (2) that if we were to select children in need of help by using the risk-crisis monitoring system, we could in fact provide preventive help. The problems with evaluating such a system are enormous because the occurrence of what one wishes to prevent is often twenty years down the road. Even when we can distinguish such high-risk groups, as with the effective use of the Gleuck scale (Meyer, Borgatta, and Jones, 1965) to delineate pre-delinquents, we must still deal with the issue of using technologies that emphasize individual construct manipulation vs. the influence of the social context.

The commitment of social work practice to the individual has also tended to keep the field focused on micro-interventions which in effect may turn out to be an inefficient use of manpower. From the perspective of primary prevention the recent work of Howard Brenner (1973) on the relationship between the job employment index, a major economic indicator, and mental hospital admissions over the past one hundred fifty years demonstrates that major efforts directed at increasing employment opportunities may be most crucial in preventing the onset of severe mental disorder. Brenner (1973) has found that an inverse relationship exists between the job employment index and the rate of hospital admissions; that is, given a short lag time, a national decrease in the availability of jobs leads to an increase in admission rates to mental hospitals.

Secondary prevention in the mental health field is identified with the notion of shortening the duration of a disorder. The classic example of secondary prevention is the U.S. Army’s efforts to deal with combat neurosis by means of the provision of psychiatric treatment services as soon as possible after onset with the goal of terminating the disorder as soon as possible. Another example of secondary prevention is the observation that prompt treatment has helped children with school phobias to resume normal school careers and prevent chronicity (Rodriguez, A., Rodrigues, M., and Eisenberg, L., 1959; Coolidge, J. D., Wilier, M. L., Tessman E. and Waldfagal, 1960). If disorders are truly more tractable in their early stages, secondary prevention becomes a strong contribution to cost effectiveness.

The problems associated with tertiary prevention have long been the primary concern of the social work profession. One of social work’s primary functions in the mental health field has been the provision of aftercare to the previously hospitalized patient. This task has largely been carried out within the framework of individual treatment. Adopting a tertiary prevention approach would change the focus of social work service efforts. In one example, Gruenberg (1966) points to the 50% reduction in the rate of onsets of severe, long-term disability associated with residence in a psychiatric hospital through the reorganization of hospital services into geographically-based clinical teams. While this observation may be attributed to other causal factors not controlled in Gruenberg’s interpretation of the results, the reorganization of the hospital was completed without the input of additional service personnel and was evaluated on the basis of its group effects.

An important point regarding attempts to do cross-area comparisons for evaluative purposes is that such comparisons should only be made with respect to changes in rates over time and these are, at best, only tenuous indicators of the effectiveness of service input. The tenuous nature of these rate changes is due to the large number of confounding variables present in the different environments that are compared.

To support adoption of a preventive public health model in social work is not to advocate that we stop caring about individuals. However, within the present framework of social values there is possibility for using limited social work resources selectively to achieve maximum results.

4. Individual Constructs vs. Social Context as a Factor in Social Influence

The next two issues — individual constructs vs. social context, and practice art vs. science — turn on questions of how people perceive the social work process. The nature of the individual constructs vs. social context issue can be most concisely stated by the observation that social work as a profession has moved away from Mary Richmond’s model and has taken up the challenge of social problems on the basis of a technology that attempts to manipulate personal constructs rather than social context. In speaking of personal consructs we refer to George Kelley’s (1955) observation that each individual develops a set of constructs to interpret his own reality and that the focus of the profession has largely been on helping the individual cope with his perception of reality. Thus, in large part, the focus has been on reordering and restructuring personal constructs.

This approach leads to a somewhat naive consideration of actual treatment efforts and proposals for service delivery. This can be illustrated by an example of planning for service to the chronically mentally ill. California has recently moved toward phasing out its mental hospital system. Aviram and Segal (1973) have pointed out that some of the slack in the system may be taken up by increased use of penal code commitments of former patients to the jails. Recently, in working on a task force on Mental Health and Justice, I learned that the group’s first task was to consider a proposal to provide direct treatment in the jails. From the outset, the group considered the proposal solely on the merits of the input of the helper/client relationship. The idea that acceptance of such a programme in the jails might lead to the legitimation of prisons as mental health facilities, perhaps in lieu of the state hospital, was not considered. The question of whether the jail is an appropriate context for helping was only considered as a result of the vocal efforts of an ex-prisoner on the task force.

Emphasis on individual constructs may explain, in part, failure to demonstrate desired outcomes. The social environment, frequently stands in the way of generating change. For example, Merton (1957) and Cloward and Ohlin (1960) point to the influence of structurally determined opportunity systems as one of the causes of delinquency. That is, they attempt to show how the social structure acts as a barrier to the delinquent’s achievement of desired goals. Similarly, in the welfare area, the nature of the labour market is a major constraint on any effort to move large groups of people off the welfare rolls with the single input of problem solving skills.

From a technological perspective one possible solution to this issue is the classification of social work intervention strategies according to the intransigence of the environmental context in which they occur. That is, social work approaches would be classified according to the extent to which the outcome of intervention can be expected to be strongly influenced by environmental contingencies. In reviewing the social work outcome literature Segal (1972) points out that evaluative attempts can be classified according to whether they deal with one or another group of clients: a population seeking treatment for psychologically based problems, or a population receiving treatment for social problems. Segal (1972) noted that those individuals making up the social-problem population seemed to be a non-voluntary population. Given this current state of our empirical understanding of the field, service efforts might be classified along the dimensions of clients’ attitude toward participation and the degree of intransigence of the problem. The possibilities that arise from interaction of these two variables are presented on Figure 2. It might be pointed out with respect to Figure 2 that whether the client makes a voluntary entrance into the service relationship may be one of the most important factors influencing his potential for a restructuring of his interpersonal system.

Figure 2.

Classification of Social Work Service Efforts as a Function of their Context and the Clients’ Attitude Toward Participation in the Service Relationship

The differing types of situations described in Figure 2 can serve as a guideline to identify different outcomes and perhaps provide some measure of the handicap a service team experiences in their service efforts. These situations might be very useful, for example, in work with an evaluating body or discussing the input of social work services for public policy purposes. In the case of the non-voluntary client the service contract is usually a second party agreement as opposed to the first party arrangement of the voluntary client. In the former situation the individual participates because he is compelled to due to his current status (e.g., prisoner).

While the relative intransigence of the service context will vary from individual to individual and there are those who will argue that the family has a high degree of intransigence, the high intransigence contexts should be viewed as extremes — e.g., the prison environment or the poverty pocket with a many-generation welfare heritage. In the case of the context with a high degree of intransigence, the contract is “limited” by the degree to which ancillary inputs such as vocational training and income are necessary to meet service goals. This is contrasted to the “full” contract — entered into in an environment with low intransigence — where the worker expects to be able to meet service goals through the use of the therapeutic technology alone.

Cell one in Figure 2 describes the “first party full-contract-helping relationship” and is best illustrated by clients voluntarily seeking help to cope with their psychological problems relating to their interpersonal relations. Clients in this situation have change options open to them which can have significant impact on the problem. In this situation clients want to change and the environment will permit such change.

Cell two refers to the “first party limited contract helping relationship.” An example - of such a situation might involve the welfare mother requesting help for severe headache — diagnosed as a psychophysiological reaction induced by the stresses associated with her economic condition. Given that she is currently receiving all the economic support available to her and the demands of her home situation will not allow for the input of additional economic resources, the intransigence of the environment limits the contract to one of social and physiological support rather than the alleviation of the problem’s cause.

The situation where there is a relatively low degree of intransigence in the service context and the client is involuntary, designated in cell three, leads to a “second-party full contract helping relationship.” Such a relationship is perhaps best illustrated by the child entering a treatment situation at the behest of his parents, where the treatment goals can be fully achieved with available therapeutic technology (e.g., the school phobia).

Finally, cell four — “The second party limited contract” — is best illustrated by the situation of the prisoner with very real psychological problems relating to his confinement, coming to a social worker at the request of the prison administration. Certainly the extent of the helping contract is limited by the inability of the social worker to bring about major changes in the prison environment in his therapeutic role and by the involuntary participation of his client. In such situations the social worker should begin his contract with his client with a straightforward confrontation of the nature of the circumstances under which the helping relationship is entered into and should be quite modest in his outcome expectations.

5. Practice Art vs. Science

The conflict between practice art and scientific endeavor is probably the most traditional issue in the service professions. Social work practice is an art and should be taught as such. This, however, does not preclude the input of science as a basis for the development of sound technologies in the field. At the centre of the art vs. science dilemma is the scientific requirement of replicability of results and consistency of method which is in direct contrast to the notion of artistic achievement. The essential character of this issue and the conflicts that have resulted from it has lead to major advances in both practice and its scientific foundations.

In his insightful article on innovations in medicine, Joseph Ben-David (1959) describes how the differential development of the professional scientific community in relation to organized medical practice in Germany and the U.S. contributed to major scientific advances made during the late nineteenth and early twentieth centuries.

The academic scientists sharply separated from the medical practitioners in Germany in the late nineteenth century. In the U.S. a more integrated type of practitioner-scientist developed. Ben-David notes that as a result of the science/practice split in Germany there was less receptivity in the sciences toward innovation. He points out that the:

… analysis of the beginnings of bacteriology and psychoanalysis lends general support to the proposition that contact with practice may be important in reorienting research toward the investigation of new and fruitful problems. The practitioner-scientists appear as forerunners, supporters, and disciples in the history of two innovations, bacteriology and psychoanalysis, the central figures in both were “role hybrids” who were led to the innovation by an abrupt change from theoretical research to applied science. (Ben David, 1959, p. 567).

Ben-David further notes that once the innovations in bacteriology became an accepted part of academic science — an event which happened later in Germany than in other countries not experiencing a severe science/practice split — the scientifically oriented methodology of the academic community enabled the Germans to quickly add to the initial findings in the field.

In social work the practice/science split began to develop with the beginning input of social and behavioral science knowledge into the field in the late 1940’s and early 1950’s. While the practice/science split has broadened in recent years the early emphasis on the doctoral-trained practitioner scientist (i.e., having doctoral training follow upon a significant amount of practice experience) tended to minimize its initial impact. However, the more recent trend in advanced social work education of training individuals with minimal practice experience for teaching and research positions in the field may lead to a social work profession that more approximates the German medical profession of the late nineteenth century.

In a manner similar to the German system described by Ben-David the first major innovation in social work practice (i. e., the introduction of psychoanalytic concepts) came from the practitioner-scientist. The profession was practice centered at the time. The introduction of psychoanalytic concepts into social work practice after WWI constituted an attempt to introduce a scientific basis into the profession.

Unlike the German system the most recent innovation in social work practice (i.e., behavior modification technologies) has come from the scientific laboratory, from the work of Pavlov, Skinner, and Thorndike. To a great extent this seems to reflect the relatively increased focus of the profession on the academic and an increased priority on the scientific. In moving in this direction the profession must continue to be concerned with practice efforts lest we lose the input of innovations from the area.

Summary and Conclusion

There are no simple solutions to the issues considered in this paper. Very likely this is because these issues represent competing dimensions underlying social work practice in particular and professional practice in the helping professions in general. It would seem that conflict generated by these issues serves as a basis for continued knowledge building in the social work field.

In this paper I have tried to provide a direction for the resolution of these issues as they confront current practice needs.

In looking at the outcome vs. the process orientation in social work, the problems of goal displacement and the bootlegging of unsupportable goals have been emphasized. The outcome-oriented issue of demonstrated effect vs. need fulfillment has brought to light the difficulties of evaluative efforts focused on the assessement of a predictable outcome for a given treatment intervention—difficulties experienced both by the practitioner and the researcher.

Cost effectiveness vs. individual commitment, the second main outcome issue, has been considered from the perspective of how the profession might best deal with the problems it has committed itself to solving in view of its limited amount of resources.

Finally the conflict between practice art and science has been viewed as productive of scientific innovation. The caution, however, has been raised that a deepening split between the academic and the practice community could result in the loss of the benefits of cross fertilization of ideas.