In social work circles client withdrawal from a treatment program commonly has been labeled as "discontinuance". Discontinuance rates have been inordinately high for both casework and group work endeavors, ranging in some instances to 59Z of all clients following the first interview (Aronson and Overall, 1966; Empey and Erickson, 1972; Goldstein, Heller, and Sechrest, 1966; Levinger, 1960; Overall and Aronson, 1963). Discontinuance represents an obvious and essential concern for social work for one overarching reason, to wit, treatment interventions cannot be implemented should the client(s) withdraw from the therapeutic relationship. Additionally, as some investigators have shown, discontinuance represents a focal concern for evaluative research since valid estimates of treatment success cannot be obtained unless early discontinuers are regarded as instances of treatment failure (cf. Lerman, 1968; Empey and Erickson, 1972).

A variety of reasons have been posited for the high rates of discontinuance in social work and allied treatment professions. Among the foremost are inaccurate or incongruent role expectations held by the two people most central to the therapeutic relationship, namely, the therapist and the client (Frank, 1961; Freeman and Simmons, 1958, 1959; Garvin, 1969; Goldstein, 1966; Heine and Trosman, 1960; Kadushin and Wieringa, 1960; Lefton, et al., 1962; Mayer and Timms, 1970; Mechanic, 1961; Olsen and Olsen, 1967; Oxley, 1966; Shapiro, 1971). These and other factors have been posited to result in counter-therapeutic client uncertainties (Erikson, 1957), anxieties (Dibner, 1967), and misperceptions (Sapolsky, 1965; Thomas, et al., 1955) and, consequently, in discontinuance. Interestingly, the great majority of studies concerning discontinuance in social work have focused solely upon endogenous features of the therapeutic relationship per se, that is, upon social variables that emanate from the interaction between client(s) and therapist(s) and that determine their ongoing interaction. For the most part, however, exogenous determinants of that relationship have been ignored in the literature. This is especially unfortunate in the case of group work since the course of treatment may depend upon a variety of socio-cultural attributes and behaviors that the various members bring to the treatment group.