Ah Beh Res. Thor. Vol. 6, pp 83-9X. 1YX.l 014&640?184 $0.00 + .

50
Prmted in Great Britain. All rights reserved. Copyright 0 1984 Pergamon Press Ltd

SELF-MANAGEMENT THERAPY FOR

DEPRESSION*

Lynn P. Rehm
University of Houston, TX, U.S.A

The aim of this paper is to review applications of models of self-management or

self-control to cognitive and behavioral therapies for depression. Self-manage-
ment tactics employed as secondary adjuncts to other forms of therapy will be
reviewed along with therapy programs employing self-management strategies as the
primary form of treatment. For the most part these approaches have been applied
with adult, primarily female, depressed outpatients. Exceptions to this rule will
be noted and issues of generality of these approaches to other patient groups will
be addressed.

Self-control concepts can serve as a model for understanding the phenomena of

depression. Theoretical models of self-control can serve both as a means of
organizing and directing research into depressive psychopathology and also as a
framework for evaluating and targeting specific depressive behaviors in psycho-
therapy. The utility of the self-control model for depression in organizing
psychopathology research is reviewed elsewhere (Rehm, 1982a). The focus of this
paper will be to review the applications of self-control to psychotherapy for
depression.

Kanfer (1970, 1971; Kanfer and Karoly, 1972) developed a model of self-control
which assumes that when individuals become aware that their behavior in a partic-
ular realm is not producing the desired or expected positive reinforcement, that
then they engage in a three-stage sequence of behaviors which act as a feedback
loop to modify the behavior in question. The first stage of Kanfer's model is
self-monitoring. In this stage, individuals focus attention on the particular
class of behavior and observe it in more or less systematic fashion. The behavior,
its antecedents, its concommitant and its consequences may be observed. The
second stage of Kanfer's model is self-evaluation in which individuals compare
the information which they have self-monitored to an internal standard and make a
judgement of the degree to which the behavior meets that standard. A modification
to this model (Rehm, 1977) adds a self-attribution component. The assumption is

*The author's research was supported in part by NIMH grants ROl MH27822 and ROl
MH34204. Earlier versions of the paper were presented at the annual meeting of
the Associations for the Advancement of Behavior Therapy, Washington, DC,
December, 1983 and, in Spanish, in Revista Latinamericana de Psicologia, 1984.
Copies of the self-control program therapist's manual can be obtained for $5 US
from Lynn P. Rehm, Ph.D., Psychology Department, University of Houston, Houston,
Texas, USA.

x3
 L. P. Rehm
that self-evaluation does not occur unless the person attributes the cause of
the behavior to him or herself. That is, the person must believe that he or she
has responsibility for and control over the behavior. Behavior over which the
person has no control is neither praise worthy nor blame worthy. Only behavior
for which one is responsible is judged at good or bad. Based on the outcome of
the self-evaluation, the final stage in Kanfer's model consists of self-reinforce-
ment. Behavior which is judged to have met the standard may be rewarded by the
individual and behavior which does not meet the standard may be punished. It is
assumed that people can modify their own behavior in the same way in which one
person might modify the behavior of a second person through the manipulation of
reward and punishment contingent on specific classes of behavior. These rewards
and punishments may be covert, (positive or negative self-statements) or they
may be overt (pleasurable or unpleasurable activities). Efforts towards long
range goals may thus be reinforced by self-administered means even when external
reinforcement may be lacking for the current behavior.

This model has a number of features which makes it adaptable to considering depres-
sive behavior. The model permits consideration of significant individual dif-
ferences in the style or manner in which individuals behave at each of the stages
of self-control. That is, there may be differences in the nature and skill with
which people self-monitor, self-evaluate, make attributions of causality, and
self-reinforce. Deficits in these various behaviors may be related to specific
forms of psychopathology. As a multistage model, the self-control model offers
the possibility of examining multiple deficits which may be significant in depres-
sion in an interrelated framework. Self-control is concerned with behavior which
people engage in when they perceive a change in reinforcement contingencies or
available reinforcements. Thus, the model may be particularly applicable to
situations which have been linked to depression, such as loss or perception of
loss of reinforcement. Lewinsohn and his colleagues (e.g. Lewinsohn and Arconad,
1981) view depression specifically as a response to a loss or lack of response
contingent positive reinforcement. To the degree that individual differences in
self-control behavior may describe individual ability to adapt to such losses,
the model may be applicable to the descriptions of how people develop or avoid
developing depression. Since self-control skills interact with situations of loss,
the model can be seen as describing depressive risk or vulnerability. Important
reinforcements in life are usually delayed and they require persistent effort to
obtain them. Kanfer's self-control model deals with ways in which individuals
supplement or replace immediate external reinforcement with self-reinforcement so
as to gain long term goals. Again, the model is applicable to issues in depres-
sion. Behavior aimed at long term goals without immediate environmental support
is often the first to diminish as the person becomes depressed. In the social
learning tradition, the self-control model encompasses both overt and covert
aspects of behavior. As a model, it has the potential for combining important
aspects of prior behavioral and cognitive models of depression.

A SELF-CONTROL MODEL OF DEPRESSION

Based on these assumptions, a self-control model of depression was proposed

(Rehm, 1977). The model hypothesized that the important core behaviors in depres-
sion could be described in terms of deficits in the various stages of the self-
control sequence. Specifically, the self-monitoring behavior of depressed persons
can be characterized by 1) a tendency to attend selectively to negative events
to the exclusion of positive events, and 2) a tendency to attend selectively to
immediate as opposed to delayed consequences of behavior. The self-evaluation of
depressed persons is characterized by 3) setting stringent self-evaluative cri-
teria for performance, and 4) making self-attributional errors consistent with
expectations of negative outcome. In the self-reinforcement phase of self-
control, depressed persons 5) self-administer insufficient contingent reward, and
6) self-administer excessive punishment. While research since 1977 suggests some
directions for modifications and additions to the self-control model of depression
 Self-Management Therapy

(cf. Rehm, 1982), in general the model continues to serve well as an organizing
framework for research and for a program of therapy intervention.

The major symptomatology of depressed persons reflects the various deficits in

self-control behavior. Selective attention to negative events is reflected in
the pessimistic outlook and distorted evaluation of experience manifested by
depressed parsons. Selective attention to immediate events is related to the
tendency of depressed individuals to be self indulgent and to have particular
difficulty in working toward long range goals. Stringent self-evaluative cri-
teria are reflected in negative self-evaluations and low self esteem. Depressive
attributions are reflected in the helpless and hopeless beliefs of depressed
individuals. Lack of self-reward and high rates of self-punishment are reflected
in self-derogation and the predominance of negative self-statements made by
depressed persons. The covert statements are assumed to function as contingent
consequences of the persons behavior. It is assumed that these reinforcement
contingencies produce the slowed behavior and lack of motivation which depressed
individuals demonstrate in many important areas of their lives.

Elements of the self-control model parallel and draw from other behavioral and
cognitive models of depression. Beck's (e.g., Beck, Rush, Shaw, and Emery, 1979)
conception of cognitive distortion in depression is represented in self-monitoring
and in self-attributional behaviors, Seligman's (1975) concept of learned help-
lessness is partly incorporated in the self-attributional component of the self-
control model. Lewinsohn's behavioral model (e.g. Lewinsohn and Arconad, 1981)
is supplemented by the self-control model in that it assumes that both self-
administered as well as externally administered reinforcements may be influen-
tial in producing depressive overt behavior.

SELF-CONTROL TECHNIQUES AS ADJUNCTS TO THERAPY FOR DEPRESSION

Self-control or self-management techniques have been used in therapy programs for

a wide variety of problems (cf. Thoresen and Mahoney, 1974; Goldfried and Merbaum,
1973; Karoly and Kanfer, 1982). Many of these procedures have been applied to
problems of depression as adjunctive procedures. In these instances the depres-
sive problem was not defined in self-control terms, but self-control techniques
ware used as a methodology for at least a portion of the therapy program.

Some of the earliest applications of self-control techniques as part of therapy

programs for depression involved attempts to modify depressive cognitions, defined
in terms of excessive negative self-statements and infrequency of positive self-
statements (Johnson, 1971; Mahoney, 1971; Tharp, Watson, and Kaya, 1974; Todd,
1972; Wanderer, 1972; Vasta, 1976). Positive self-statements were to be prompted
by frequent activities or practiced contingently following accomplishment of
target behaviors. Jackson (1972) reported a case study involving a fairly com-
prehensive self-control treatment program for depression. In this instance, the
patient's depression was conceptualized in self-control terms. Deficient self-
control behavior was remediated by instructions on how to set realistic goals for
task performance, how to develop a self-reinforcement program for successful task
completion, and how to self-monitor performance accurately. The patient reported
a decrease in depression and a degree of generalization of the self-control
behavior to other aspects of her life. A two-month follow-up indicated mainten-
ance of treatment gains.

Two group design studies have been reported which used a variety of techniques
with overall self-control rationales. Hilford (1975) compared a self-control
package to nondirective group therapy and a no-contact control condition in a
study with hospitalized depressed patients. The experimental condition encouraged
self initiated behavior change by presenting six suggestions. 1) Complete even
the smallest task to the best of your ability, 2) think positively and try to be
constructive, 3) be aware of your effect on others, 4) become familiar with other
 L. P. Rehm
people, 5) get involved in activities with other people and 6) stop negative
thoughts by thinking of pleasant events or by instructing yourself to relax.
Experimental subjects improved significantly more than control subjects on self-
reported depression and social adjustment including ward behavior ratings. These
results were maintained at follow-up. Hedlund and Thompson (1980) used an approach
which they described as an educational self-management format to teach elderly
patients how to control their depression. This condition was compared with indi-
vidual behavior therapy condition. In both instances, therapy targets were drawn
from Lewinsohn's behavioral approach to depression. Both approaches prcduced
significant change on measures of depression, loneliness, and hopelessness with
no differences between the two conditions. The authors conclude that the educa-
tional self-management approach has many practical advantages for treating depres-
sion in the elderly.

It should also be noted that there are many self-control or self-management proce-
dures which are incorporated as part of depression therapy programs with very
different theoretical rationales. For example, Peter Lewinsohn's behavioral
approach (e.g. Lewinsohn and Arconad, 1981) relies heavily on self-monitoring
of pleasant events. In this approach, self-monitoring is seen primarily as an
assessment modality rather than as an intervention per se. Based on the self-
monitoring information activities are scheduled. This is a self-evaluation
procedure in that it involves setting explicit, limited and realistic criteria
for behavior change. The program uses a variety of reinforcement contingencies
for encouraging increases in pleasant activities, including a sophisticated self-
reinforcement program involving an accumulation of points which can later be
cashed in for rewards. Self-reward and self-punishment are also used to control
covert and overt behavior.

In the most recent form of Lewinsohn's behavioral program (Brown and Lewinsohn,
1980) the various therapy strategies or modules have been incorporated into a
psychoeducational course which teaches patients how to cope with their depres-
sion. In essence, this has moved this program into a self-management format,
Brown and Lewinsohn demonstrated that the approach could be just as successful
when subjects used the psychoeducational text by themselves with minimal therapist
consultation.

Self-control techniques are also used in Beck's cognitive therapy (Beck, Rush,
Shaw, and Emery, 1979). Among the many specific techniques of this approach
are self-monitoring assignments whereby the patients may keep records of a
variety of specific activities. In this approach, self-monitoring is used both
as an assessment procedure and as an intervention in the sense that data are col-
lected to counter unrealistic distortions of weekly events. Patients are also
taught to self-monitor negative self-statements or what Beck refers to as "auto-
matic thoughts". Self-evaluation techniques are involved in Beck's use of
graded task assignments. The program is explicitly aimed at countering perfec-
tionism or high standard setting in depression. Beck's program also involves
explicit exercises to modify self-attributions in order to help patients deal
with excessive self-blame. Although contingent self-reinforcement is not explic-
itly part of Beck's program, attempting to counter depressive thoughts with
positive self-statements may, in fact, be contingent positive self-reinforcement.
Self-control techniques are also major portions of other complex behavioral and
cognitive programs. McLean (McLean, Ogston, and Grauer, 1973; McLean and
Hakstian, 1979) describes complex therapy programs which emphasize teaching
patients social learning principles so that they can develop and implement their
own self-managed changes programs. Dunn (1979) described a cognitive modifica-
tion program involving having patients keep a "self dialog diary" as a means of
helping them make more positive self-evaluations and more adaptive self-attribu-
tions regarding life events.

For the most part, these programs have used self-management strategies as adjuncts
or as elements of complex therapy packages with diverse rationales and theoretical
 Self-Management Therapy

framesworks. The next set of studies to be reviewed describe a therapy program

derived from the self-control model of the nature of depression.

SELF-CONTROL THERAPY PROGRAM FOR DEPRESSION

The self control program for depression is premised on the idea that depression
can be described as a series of deficits in self-control behavior (Rehm, 1977).
If this model holds, then an appropriate therapeutic strategy would be to attempt
to modify each of these potential deficits in sequence. It is assumed that
depressed individuals may vary in which deficits are particularly salient in
their depression but that attention to each phase in sequence would remediate
most depressions. The therapy program is highly structured with a strongly didac-
tic flavor. The program aims at teaching self-control concepts relevant to each
potential deficit by presentations, discussions, paper-and-pencil exercises, and
homework assignments. The program itself has been described elsewhere (Rehm,
1981a). For our research purposes, the program has been offered in a group format
with weekly, one and one-half hour meetings. In various studies the program has
varied from six to twelve weeks in its presentation. As we presently use it, it
is a ten week program. For the purposes of research, the subject population has
been symptomatic community volunteers. Announcements have been made in various
news media that a research program is available for women between the ages of
18 and 60 who feel that they have a significant problem with depression and who
would be willing to volunteer for the therapy program in a research context. Our
studies have used women subjects because research tends to show that depression
is more frequent in women and research practicalities favored homogeneous female
groups. Interested volunteers call in to a university based clinic for assess-
ment and screening appointments. The volunteers are screened on stringent cri-
teria including paper-and-pencil self-report assessments of severity of depres-
sion (Beck Depression Inventory, Beck, Ward, Mendelson, Mock and Erbaugh, 1961;
and/or the MMPI-D, Hathaway and McKinley, 1967), and on diagnostic interview
criteria. In our latter studies the Research Diagnostic Criteria (Spitzer,
Endicott and Robins, 1978) were employed with the structured Schedule for
Affective Disorders and Schizophrenia (Endicott and Spitzer, 1978). Only about
one of eight persons who inquire about the program meets all screening criteria
and begins the program. The resulting research samples are representative of
moderately to severely depressed outpatients with nonpsychotic, nonbipolar
Major Affective Disorders. On the average, they are in their late 30s and are
somewhat above average in socio-economic status though a wide range of education,
occupation and income status is represented. Most report prior episodes of
depression and about half have had prior psychotherapy. To be eligible for the
study, none of the subjects could be currently taking major tranquilizers or
antidepressant medications. Around 15% typically have prescriptions for minor
tranquilizers used in nonabusive fashion.

Assessment

In each of the studies, assessment of outcome has been accomplished by a battery

of assessment techniques. In all cases, self-report of depression has been
assessed by the Beck Depression Inventory (Beck, Ward, Mendelson, Mock and
Erbaugh, 1961) and the MMPI-D (Hathaway and McKinley, 1967). In the later four
studies, clinician rating scales have also been used to assess depression.
These have been the Hamilton Depression Rating Scale (Hamilton, 1964) and the
Raskin Three-Item Depression Rating Scale (Raskin, Schulterbrandt, Reatig, and
Rice, 1967). General psychopathology has been assessed by self-report (e.g. full
MMPI or SCL-90R; Derogatis, 1977), and by clinician ratin&s on the Global Assess-
ment Scale (Spitzer, Gibbon and Endicott, 1973). Specific self-control attitudes
and beliefs have been assessed by a paper-and-pencil questionnaire devised for
this research program. Finally, the studies have employed a series of experimenta
measures of nonverbal, paralinquistic and verbal content codes assessed from
 L. P. Rehm
pre- and posttherapy interviews and from early and late therapy sessions. A
summary of the results on the major depression variables for the six outcome
studies is shown in Tables 1 and 2.

Validation Studies

The first study in the series (Fuchs and Rehm, 1977) compared a six week version
of the program to nonspecific group therapy and waiting list conditions, with 8,
10 and 10 subjects in the three conditions respectively. On both the BDI and
the MPE'I-D the waiting list group showed essentially no improvement, the non-
specific group moderate improvement, and the self-control therapy the most
improvement both at posttest and at six-week follow-up. In terms of absolute
criteria for improvement (below 11 on the BDI and below 70 on the MMPI-D) all the
self-control subjects were within the normal range whereas only 3 of the 10 non-
specific subjects were in this range. The self-control condition also produced
the greatest reduction in general psychopathology (total MMPI elevation) and
these subjects showed greater increases in activity level on a therapy group
interaction measure and on a paper-and-pencil pleasant activity questionnaire.
Self-control subjects also showed the greatest changes on experimental paper-
and-pencil measures of self-control attitudes and beliefs.

Given the encouraging results of the first study, it was decided to evaluate
the program in a second study against a more stringent comparison condition
(Rehm, Fuchs, Roth, Kornblith and Romano, 1979). Assertion training was chosen
since it has been suggested as an appropriate therapy for at least a subset of
depressed patients (e.g. Wolpe, 1971). Assertion training has frequently been
a part of larger behavioral therapy packages for depression. Fourteen subjects
completed the self-control program and 10 the assertion program. On the BDI and
MMPI-D, there was relatively little improvement in the six weeks between screen-
ing and pretherapy. Both groups improved from pretherapy to posttherapy with
greater improvement for the self-control condition. Therapy gains were main-
tained at six weeks follow-up. The content of the two therapies was validated
independently in that assertion subjects showed greater improvement on the
Wolpe-Lazarus Assertion Scale (Wolpe and Lazarus, 1967) and the self-control
subjects showed greater improvement on our experimental self-control question-
naire. Behavioral observation measures during therapy and the activity ques-
tionnaire also indicated greater improvement for the self-control condition.

These first two studies were carried out in close temporal proximity. Follow-up
data on the combined studies were reported by Romano and Rehm (1979). The self-
control subjects from the first two studies were combined (n=16) and compared
with the available subjects in the assertion (n=6) and nonspecific therapy
(n=7). At one year, there were no longer any significant differences between
conditions on the BDI and MMPI-D scales with all groups having improved over
pretherapy levels. However, a significantly greater proportion of the self-
control subjects stated that they had not experienced episodes of depression
during the follow-up year and nearly half of the subjects in the nonspecific
and assertion conditions had obtained additional psychotherapy during the year,
whereas none of the self-control subjects had done so. The latter results indi-
cate to us that the self-control subjects had indeed acquired skills which
helped them to cope with situations and thus avoid additional episodes of depres-
sion on their own in contrast to the comparison therapy subjects who required
additional therapy in order to show improvement at the end of one year.

Disassembly Studies

Given the success of the therapy in the first two studies, it was decided to
attempt to assess the efficacy of the major components of this complex package
program. Thus, the third study of the series (Rehm, Kornblith, O'Hara, Lamparski,
 Self-Management Therapy 89

TABLE 1 Comparison of Mean Self-Report Depression Scores for

Self-Control Therapy and Other Treatments and Control Conditions

Beck MMF'I-D

N Pretest Posttest Pretest Posttest

Study and Condition Score Score Score Score

Fuchs & Rehm (1977)

Comprehensive self control 8 21.4 4.7 85.3 59.9

Non-specific group therapy 10 23.6 14.3 83.9 77.4
Waiting list control 10 23.2 21.4 80.9 81.6

Rehm, Fuchs, et al. (1979)

Comprehensive self control 14 21.3 6.0 83.4 76.1

Social skills training 10 20.3 12.5 82.4 62.0

Rehm, Kornblith, et al. (1981)

Comprehensive self control 10 29.4 19.1 86.8 80.0

Self monitoring & self
evaluation 12 28.4 18.5 88.6 75.7
Self monitoring & self
reinforcement 12 26.4 18.0 88.0 73.7
Self monitoring only 9 27.9 11.6 85.2 74.8
Waiting list control 15 26.0 21.9 88.3 84.3

Kornblith, Rehm, et al. (in press)

Comprehensive self control 11 29.3 17.6 86.0 71.3

Self monitoring & self
evaluation 12 28.4 15.5 82.8 76.9
Self control principles
only 11 27.5 10.9 78.5 68.3
Dynamic group psycho-
therapy 5 25.4 3.6 77.4 66.6

Rehm, Lamparski, et al.

Combined self control a 27.5 10.9 85.6 58.6

Behavioral self control 13 32.7 10.5 85.7 63.3
Cognitive self control 11 28.4 13.8 85.7 67.2
Waiting list 12 21.9 21.9 82.3 79.5

Rehm, Kaslow, et al.

Combined self control 34 30.2 11.0 86.8 71.0

Behavioral self control 35 28.9 10.4 82.5 69.0
Cognitive self control 35 28.9 10.5 82.0 66.0
90 L. P. Rehm

TABLE 2 Comparison of Mean Clinician Depression Scores for Self-

Control Therapy and Other Treatments and Control Conditions

Hamilton Raskin

N Pretest Posttest Pretest Posttest

Study and Condition Score Score Score Score

Rehm, Kornblith, et al. (1981)

Comprehensive self control 10 16.0 9.0 8.8 5.8

Self monitoring & self
evaluation 12 15.0 10.9 8.5 5.6
Self monitoring & self
reinforcement 12 17.3 10.8 9.3 6.2
Self monitoring only 9 18.8 8.1 8.3 4.8
Waiting list control 15 18.9 14.8 8.9 8.5

Kornblith, Rehm, et al. (in press)

Comorehensive self control 11 17.8 9.4 9.4 6.8
Seli monitoring & self
evaluation 12 18.0 6.6 9.8 5.8
Self control principles
only 11 16.9 6.0 9.7 5.2
Dynamic group psycho-
therapy 5 14.6 3.8 9.2 4.6

Rehm, Lamparski, et al.

Combined self control 8 15.0 3.7 8.1 3.8

Behavioral self control 13 18.2 7.6 8.8 5.5
Cognitive self control 11 19.5 8.4 9.8 5.8
Waiting list 12 18.4 15.6 8.6 8.3

Rehm, Kaslow, et al.

Combined self control 34 18.3 6.5 9.1 5.5

Behavioral self control 35 18.2 7.3 9.9 5.4
Cognitive self control 35 18.4 6.9 9.2 4.7
 Self-Management Therapy
Romano, and Volkin, 1981) compared the full self-control program (n=9) to a ver-
sion of the program eliminating the self-evaluation component (n=l2), a version
of the program eliminating the self-reinforcement component (n=ll), and a ver-
sion of the program eliminating both the self-evaluatjon and self-reinforcement
components (n=9). An additional 15 subjects served as a waiting list control
during the seven weeks of the therapy program. In comparison to the waiting
list group, all four therapy conditions produced consistent evidence for a treat-
ment effect on self-report and clinicianmeasures of depression and on the self-
control and activity questionnaires. No consistent differences were found how-
ever, among the experimental groups. Thus the study did not provide evidence for
an independent contribution to the program of the self-evaluation or self-rein-
forcement components.

A second disassembly study was undertaken which partially replicated the first
disassembly study and added another dimension of disassembly (Kornblith, Rehm,
O'Hara, and Lamparski, 1983). In this study, the full self-control program
(n=ll) was compared to a version of the program omitting the self-reinforcement
component (n=12) as in the last study, and a version containing all of the
didactic presentations but eliminating the explicit homework assignments. Two
therapists saw one group in each of the three conditions. A third therapist
saw an additional five subjects in one group in an interpersonally oriented com-
parison condition. The length of therapy in this study was increased to 12
weeks because we felt that the material we presented to the participants was
becoming more complex and, therefore, needed to be presented at a slower pace.
The overall therapy effects were comparable to those in the first disassembly
study. That is, there were no significant differences among any of the condi-
tions on any of the major dependent variables assessing depression,

The results of these two studies were surprising to us. Anecdotally, subjects
in the program have consistently reported to us that various aspects of the
program including the self-evaluation and self-reinforcement components and
surely the homework assignments, were major contributors to their improvement.
It may be that it is difficult to detect improvement over and above that pro-
duced by the initial component of the program and that with the relatively small
n's of the studies, this was difficult to do. It may also be that the elements
of the program which we experimentally withdrew were still implicit in what the
participants did receive. For instance, defining positive activities and having
subjects monitor these may imply realistic self-evaluative criteria and self-
reinforcement for accomplishing activity goals. Concrete descriptions of depres-
sive behavior may lead to behavior change without explicit homework assignments.
Indeed, a couple of the subjects in the "no homework" condition told us that
they devised their own homework assignments.

Behavioral and Cognitive Targets

The final two studies in our series attempted to assess the interactions between
subject deficits, therapy targets, and outcome measures. In particular, we were
interested in comparing versions of the self-control program aimed at a behavioral
target, i.e. increasing overt activity level, and a version aimed at changing a
cognitive target, i.e. the content of covert statements about one's self. Three
therapy manuals were devised. The behavioral target manual focussed on monitor-
ing positive activities, developing realistic behavioral goals and subgoals, and
using pleasurable activities as self-administered rewards for completing more
difficult tasks. The cognitive target version of the program had subjects moni-
tor self-statements, develop realistic self-evaluative and attributional self-
statements, and use positive self-statements as covert rewards. A third version
of the program combined both sets of targets. In a smaller study evaluating
these alternative programs (R&m, Lamparski, Romano, and O'Hara, in preparation)
13 subjects were seen in the behavioral target condition, 11 in the cognitive
condition, 8 in the combined condition and 12 subjects served as a waiting list
92 L. P. Rehm
control. All three active therapy conditions improved to a greater extent than
the waiting list control with no differences among them. This validated the
idea that each therapy condition was indeed an active and effective therapy
strategy.

A second larger study evaluating these conditions involved 35 subjects in a

behavioral target condition, 35 in a cognitive target condition and 34 in a com-
bined condition (Rehm, Kaslow, Rabin and Willard, 1981). Once again on self-
report and clinician measures of depression all three conditions improved to a
clinically and statistically significant degree but there were no differences in
outcome among groups. It was notable that on specific therapy target outcome
measures, there were no differences between conditions. That is, the three con-
ditions did equally well on measures of activity level and of depressive cogni-
tions. Further, subjects who showed particular deficits in the area of activity
or cognition did equally well in depression outcome in each of the three condi-
tions. Thus, the therapy conditions were nonspecific as to subject deficit or
outcome target.

This latter study had a sufficient 4 to allow us to explore other questions with
regard to prediction of outcome for subgroups of subjects (Rehm, Kaslow, Rabin
and Willard, 1981; Rehm, 1982b). The problem was looked at in two ways. First,
subjects were divided into specific subgroups and comparisons were made of their
outcome across and between therapy conditions (Rehm, 1982b). Secondly, for out-
come generally an exploratory regression equation was developed to identify pre-
dictors of depression outcome (Rehm, Kaslow, Rabin and Willard, 1981). Relat-
ively few differences were found for subgroups of participants. Among demo-
graphic variables, only age produced a small effect, such that younger subjects
tended to do better. This is typical in psychotherapy research. A later age
of onset and less overall psychopathology were also related to positive outcome
as were some of the events schedule measures (less pleasure in positive activi-
ties, fewer positive self-statements, and more negative events). Higher scores
on Rosenbaum's (1978) Self-Control Scale were associated with a better outcome.
Holding positive self-control attitudes and beliefs initially was related to
success in the program. Success was also related to positive expectancy for
involvement in the program. The strongest effect was for pretest depression.
Higher initial depression scores correlated with higher posttest scores.

An aggregate one year follow-up study was also done across all of our studies
(Rabin, Rehm, and Leventon, 1983). Results confirmed those of the earlier fol-
low-ups. Beck Depression Inventory and MMPI-D scores showed a continued therapy
effect when compared to pretest scores but most differences between self-control
therapy and control therapies had disappeared. Self-control subjects on ques-
tionnaires reported fewer, shorter and less severe subsequent episode of depres-
sion and were more likely to have coped with these without the help of addi-
tional psychotherapy. These results are again consistent with the self-manage-
ment skills orientation of the program.

Additional Studies of the Self-Control Therapy Program

The self-control therapy program has been replicated elsewhere in several stud-
ies using our therapy manuals. Fleming and Thornton (1980) treated 37 volunteer
depressed subjects in what they refer to as cognitive therapy (based on Shaw,
1977) versus behavioral therapy (based on the Fuchs and Rehm self-control manual)
in comparison to a nondirective therapy condition. All conditions improved
significantly at posttest and follow-up and the authors conclude that there were
no differences between conditions although the behavioral (i.e. self-control)
condition produced the greatest improvement on several measures in their bat-
tery. Roth, Bielski, Jones, Parker and Osborn (1982) saw 26 community volunteers
in either self-control therapy or self-control therapy plus desipramine hydro-
chloride, a tricyclic antidepressant drug. On the BDI the self-control plus
 Self-Management Therapy
medication group responded faster during therapy but at posttest and at three
month follow-up there were no significant effects between conditions. There
were no differences at any point on the Hamilton Rating Scale for Depression.
Rothblum, Green and Collins (1979) conducted two small studies using the Fuchs
and R&m self-control manual. In the first, 18 obese depressed persons and in
the second study, 33 depressed persons were seen in either of two versions of
the self-control program. The first version stressed client responsibility in
selecting behaviors for monitoring and goal-setting while a second version
stressed the therapist's role in helping to set goals. Results tended to favor
the active role of the therapist which is more consistent with our original use
of the therapy program. Tressler and Tucker (1980) saw 17 female volunteer
depressed persons in a disassembly study somewhat similar to ours. They com-
pared two versions of the self-control program, one consisting of the self-
monitoring and self-evaluation components and the second consisting of the self-
monitoring and self-reinforcement components. At posttest and 12 week follow-up
the self-monitoring plus self-reinforcement condition proved significantly
superior.

There have been a number of recent reports of additional applications of the self-
control program to more varied populations. Glans and Diets (1980) advertised
the availability of the program to individuals in a community as part of the
community mental health center's outreach program. A diverse group of volunteer
participants included currently depressed, previously depressed, and spouses or
other relatives of depressed persons. In general, participants showed gains in
depression inventory scores as a result of participation in the program.
Kornblith and Greenwald (1982) gave a preliminary report on the application of
the self-control therapy program for inpatients. As with the Hilford (1975)
program, the self-control program here was an adjunct to an active medication-
oriented hospital routine. Given the relatively limited hospital stay in this
acute treatment facility, the program was presented at a more rapid pace with
two or three sessions per week. While this allowed less time to practice assign-
ments between sessions, the authors found that even within the restricted hospi-
tal environment the assignments seemed to be sufficient to teach utilization of
the principles. The frequent sessions also allowed the therapist to have closer
supervision over participants' efforts. The program was also simplified some-
what for this more rapid presentation. Preliminary indications were that the
program added significantly to the depression outcome of patients who had
participated.

Rogers, Kerns, Rehm, Hendler, and Harkness (1982) reported on the use of the
self-control therapy program with a small number of renal dialysis patients.
Male patients in a Veterans Administration Hospital renal dialysis treatment
program who were identified as having significant levels of depression were seen
individually in the self-control program and were compared to a second group of
dialysis patients who were seen in a nonspecific individual psychotherapy. The
individual application of the program followed the group manual and protocol
fairly directly. Greater improvement on depression was demonstrated for the
self-control patients as opposed to the nonspecific psychotherapy patients. It
is notable that depression improved considerably despite the relatively serious
continuing medical condition of the patients involved.

CONCLUSIONS

A number of general statements can be made about the current status of the self-
control therapy program for depression. Primary among these conclusions is
that the program has been demonstrated to be effective. Results have been
replicated a number of times in our own clinics as well as by a number of inde-
pendent investigators at other sites. Participants with clinically significant
levels of depression have improved to clinically and statistically significant
degrees and, in most cases, improved to a greater degree than subjects in
 L. P. Rehm
control or comparison conditions. Improvement has been shown on self-report
and clinician depression scales, surveys of general psychopathology, and a
variety of measures of specific depressive cognitive and behavioral factors.
Effects appear to be maintained in short and long term follow-up. Despite the
fact that many individuals who have been through the program have additional
episodes of depression during the following year, these episodes appear to be
reduced in frequency, intensity and duration. This result is consistent with
the skills acquisition model involved in self-management programs. This program
appears to be credible and acceptable to participants as well.

Our primary research population has been depressed adult women. There is some
evidence that the program has fairly wide applicability with regard to patient
groups. Several of the studies cited used males exclusively or in mixed sex
groups. In no instance was there any contraindication of the effectiveness of
the program for application to males as well as females, Our prediction and
subpopulation studies suggest that there is relatively little differential
effectiveness based on demographic variables beyond a small age effect. To
date our program has not been applied to the elderly. The Hedlund and Thompson
studies suggest that self-management approaches might be quite effective for
the elderly. A number of adaptations of the self-control program might be called
for. Similarly there have, to date, been no application of the program to child-
ren or adolescents. There have been, however, a number of applications of
specific self-control techniques to children with problems closely related to
depression, such as low esteem (cf. Kaslow and Rehm, 1983). The Kornblith and
Greenwald (1982) extension of the program to use with inpatients appears promis-
ing. It should be noted that in this instance the program is used in conjunc-
tion with medications with patients who are fully ambulatory and able to use the
program. Some severely depressed inpatients at least initially would not be
sufficiently mobilized to be able to take part in the program. Applications to
medical populations also presents an interesting area of possible future appli-
cation. Depression has been identified as a major complicating factor in the
subsequent course and recovery of patients suffering from a number of chronic
disorders or recovering from severe medical conditions. A relatively structured,
short term intervention as an adjunct treatment for such patients could be of
considerable benefit.

In our research, the therapy program has been conducted in the group format.
Clinically we have found the program quite adaptable as well for individual
therapy. In Rogers' dissertation study (Rogers, Kerns, Rehm, Hendler and
Harkness, 1982), the program was adapted fairly directly to an individual
research protocol. In the clinic when we have seen individual patients, we have
tended to be more flexible with the program. This includes spending more or
less time on particular issues as a function of the patient's ability to under-
stand and use the concepts and as a function of the apparent relevance of the
concept to the particular person's problems. Typically more time is spent in
individual applications with goal setting for particular problem areas. Often
other techniques, such as role playing, are added as a form of rehearsal for
subgoal activities. More time is usually spent after the program elements have
been taught following the client in utilization of the program.

There are a number of future directions for the program, both in terms of
research and clinical application. It is quite clear that research to date has
not been able to identify the specific active elements in this therapy program.
Among the several cognitive-behavioral therapy programs which have been recently
demonstrated to be effective with depressed persons, our program has probably
been the most extensively researched in terms of a component analysis. When
examined closely, it is possible to note that these various programs, although
their theoretical rationales are quite different, in practice have many common-
alities. Among the programs that might be cited are Beck's cognitive therapy
(Beck, Rush, Shaw, and Emery, 1979); Lewinsohn's behavioral therapy (Lewinsohn
and Arconad, 1981); McLean's behavioral marital therapy (McLean and Hakstian,
 Self-Management Therapy

1979; McLean, Ogston, andGrauer,1973) and Hersen and Bellack's social skills
approach (Hersen, Bellack, and Himmelhoch, 1980). Each of these programs share
at least four common characteristics. First, each has a clear, concrete ration-
ale. This rationale is explicitly taught to the participant such that the
,participants learn an alternative way of conceptualizing their problems which
may lead them out of the helpless and circular logic which often characterizes
patients' views of their problems. Secondly, each of these programs is relative-
ly structured. The rationales translate rather directly into cognitive and
behavioral action strategies. Third, each of these programs involves instigation
to change. Considerable emphasis is put on homework assignments to practice
aspects of the rationale in relatively structured fashion between sessions in
actual life experiences. Fourth, each of these programs provides relatively
explicit progress feedback. Clients in all these programs can see, usually in
quantitative terms, evidence of their own progress from one point to another as
they proceed through the program. Various self-monitoring records or logs, suc-
cessive activity schedules, and changes in explicit self-statements are all docu-
mented and made clear to the patient. It may be that it is these elements
rather than the validity of the theories behind the rationales which are most
important in the effectiveness of the therapy programs. To some extent, the
validity of the theories and the effectiveness of the therapies are independent
matters. Therapy research in the future might well be addressed to evaluating
these more abstract common components. Taken from this perspective, the self-
control program can be seen as teaching participants concrete self-management
skills in a program that is simply tailored to the typical problems of depres-
sion. The program can be seen as teaching self-administered behavior modifica-
tion around various problems related to depression.

While the program has been demonstrated to be relatively effective in research

contexts, information is still relatively scarce with regard to the program's use
in typical clinical situations. While the program is beginning to be used in a
number of such situations, systematic data gathering is relatively rare. The
structure and group format of the program makes it particularly adaptable to
clinical situations where it can be offered in a relatively cost efficient manner.
Its potential for low cost administration, its efficacy, and its implications
for prophylaxis needs to be evaluated in larger and more varied clinical situa-
tions.

Finally, the program needs to be evaluated vis-a-vis its use with pharmacotherapy
for depression. Pharmacotherapy is well established in the research literature
and is the modal treatment for depression available today. Research evidence on
the relatively efficacy and interaction of pharmacotherapy and the new cognitive-
behavioral methods is relatively sparse but to date evidence suggests that the
cognitive-behavioral methods are at least as effective and possibly more effec-
tive. There is some evidence that dropouts are fewer and maintenance is better.
Pharmacotherapy may produce a more rapid initial effect, may be more effective
for neurovegetative symptoms, and might be more effective for the severest cases
but evidence is still scanty, The field badly needs data, not just on the com-
parative efficacy of cognitive-behavioral and pharmacological interventions, but
on the effectiveness of their interaction and on the effectiveness of each pro-
gram for the therapeutic failures of the other types of programs. Would the
cognitive-behavioral programs be effective for tricyclic antidepressant nonre-
sponders, and would tricyclics be effective for cognitive and behavioral therapy
nonresponders? The evidence to date suggests at least that the cognitive-
behavioral therapy programs are a valid current alternative to drugs in treat-
ment of depression but we could learn much more about their possible interactions.
 96 L. P. Rehm
REFERENCES

Beck, A. T., Rush, A. G., Shaw, B. F., and Emery, G. (1979). Cognitive Therapy
of Depression. Guilford Press, New York.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., and Erbaugh, J. (1961). An
inventory for measuring depression. Archives of General Psychiatry, 6,
561-571.
Brown, R. A. and Lewinsohn, P. M. (1980). A psychoeducation approach to the
treatment of depression: Comparison of group, individual and minimal con-
tact procedures. Unpublished paper, University of Oregon.
Derogatis, L. R. (1977). SCL-90 Administration, Scoring and Procedures Manual-
I, Johns Hopkins University.
Dunn, R. J. (1979). Cognitive modification with depression-prone psychiatric
patients. Cognitive Therapy and Research, 3, 307-317.
Endicott, .I., and Spitzer, R. L. (1978). A diagnostic interview: The Schedule
for Affective Disorders and Schizophrenia. Archives of General Psychiatry,
35, 837-844.
Fleming, B. M., and Thornton, D. W. (1980). Coping skills training as a compo-
nent in the short-term treatment of depression. Journal of Consulting and
Clinical Psychology, 48, 652-655.
Fuchs, C. Z., and Rehm, L. P. (1977). A self-control behavior therapy program
for depression. Journal of Consulting and Clinical Psychology, 45, 206-215.
Glans, L. M., and Dietz, R. E. (1980, August). Building individual competence:
Coping with depression and fear. Paper presented at the meeting of the
American Psychological Association, Montreal.
Goldfried, M. R., and Merbaum, M. (1973). Behavior Change Through Self-Control.
Holt, Rinehart and Winston, New York.
Hamilton, M. (1964). A rating scale for depressive disorders. Psychological
Reports, l4, 914.
Hathaway, S. R., and McKinley, J. C. (1967). MMPI Manual (Revised 1962), The
Psychological Corporation, New York.
Hedlund, B., and Thompson, L. W. (1980, September). Teaching the elderly to
control depression using an educational format. Paper presented at the
meeting of the American Psychological Association, Montreal.
Hersen, M., Bellack, A. S., and Himmelhoch, J. M. (1980). Treatment of unipolar
depression with social skill training. Behavior Modification, 4, 547-557.
Hilford, N, G. (1975). Self initiated behavior change by depressed women fol-
lowing verbal behavior therapy. Behavior Therapy, a, 703.
Jackson, B. (1972). Treatment of depression by self-reinforcement. Behavior
Therapy, 3, 298-307.
Johnson, W. G. (1971). Some applications of Howoe's coverant control therapy:
Two case reports. Behavior Therapy, 2, 240-248.
Kanfer, F. H. (1970). Self-regulation: Research, issues, and speculations. In
C. Neuringer and 3. L. Michael (Eds.), Behavior Modification in Clinical
Psychology, Appleton-Century-Crofts, New York.
Kanfer, F. H. (1971). The maintenance of behavior by self-generated stimuli
and reinforcement. In A. Jacobs and L. B. Sachs (Eds.), The Psychology
of Private Events: Perspectives on Covert Response Systems. Academic
Press, New York.
Kanfer, F.-H., and Karoly, P. (1972). Self-regulation and its clinical appli-
cation: Some additional conceptualization. In R. C. Johnson, P. R.
Dokecki, and 0. H. Mowrer, Socialization: Development of Character and
Conscience. Holt, Rinehart, and Winston, New York. pp. 428-437.
Karoly, P., and Kanfer, F. H. (Eds.). (1982). Self-Management and Behavior
Change: From Therapy to Practice. Pergamon Press, New York.
Kornblith, S. J., and Greenwald, D. (1982, November). Self-control therapy
with depressed inpatients. Paper presented at the meeting of the
Association for the Advancement of Behavior Therapy, Los Angeles.
 Self-Management Therapy 97

Kornblith, S. J., Rehm, L. P., O'Hara, M. W., and Lamparski, D. M. (in press).
The contribution of self-reinforcement training and behavioral assignments
to the efficacy of self-control therapy for depression. Cognitive Therapy
and Research.
Lewinsohn, P. M., and Arconad, M. (1981). Behavioral treatment of depression:
A social learning approach. In J. F. Clarkin and H. I. Glazer (Eds.),
Depression: Behavioral and Directive Intervention Strategies. Garland
STPM Press, New York.
Mahoney, M. J. .(1971). The self-management of covert behavior: A case study.
Behavior Therapy, 2, 575-578.
McLean, P. D., and Hakstain, A. R. (1979). Clinical depression: Comparative
efficacy of outpatient treatments. Journal of Consulting and Clinical
Psychology, 9, 818-836.
McLean, P. D., Ogston, K. and Grauer, L. (1973). A behavioral alproach
. to the
treatment of depression. Journal of Behavior Therapy and Experimental
Psychiatry, 5, 323-330.
Rabin, A. S., Rehm, L. P., and Leventon, B. (1983, December). Aggregate follow-
up results following self-control therapy for depression. Paper presented
at the meeting of the Association for the Advancement of Behavior Therapy,
Washington, DC.
Raskin, A., Schulterbrandt, J., Reatig, N., and Rice, C. E. (1967). Factors of
psychotherapy in interview, ward behavior, and self-report ratings of
hospitalized depressives. Journal of Consulting Psycholopy, 3l, 270-278.
Rehm, L. P. (1977). A self-control model of deprssion. Behavior Therapy, 8,
787-804,
Rehm, L. P. (1981a). A self-control therapy program for treatment of depres-
sion. In J. F. Clarkin and H. I. Glazer (Eds.), Depression: Behavioral
and Directive Strategies. Garland STPM Press, New York.
Rehm, L. P. (Ed.) (1981b). Behavior Therapy for Depression: Present Status
and Future Directions. Academic Press, New York.
Rehm, L. P. (1982a). Self-management and depression. In P. Karoly and F. H.
Kanfer (Eds.), The Psycholopy of Self-Management: From Theory to Practice,
Pergamon Press, New York.
Rehm, L. P. (1982b). Outcome of self-control therapy for depression within
subpopulations. In P. M. Lewinsohn (Chair), Treatment of unipolar depres-
sion across the age span. Symposium presented at the meeting of the
American Psychological Association, Washington, DC.
Rehm, L. P., Fuchs, C. Z., Roth, D. M., Kornblith, S. J., and Romano, J. M.
(1979). A comparison of self-control and assertion skills treatments of
depression. Behavior Therapy, lo, 429-442.
Rehm, L. P., and Kaslow, N. J. (in press). Behavioral approaches to depression:
Research results and clinical recommendations. Child and Family Behavior
Therapy, supplement.
Rehm, L. P., Kaslow, N. J., Rabin, A. S., and Willard, R. (1981, August).
Prediction of outcome in a behavior therapy program for depression. Paper
presented at the annual meeting of the American Psychological Association,
Los Angeles.
Rehm, L. P., and Kornblith, S. J. (1979). Behavior therapy for depression: A
review of recent developments. In M. Hersen, R. M. Eisler and P. M.
Miller (Eds.), Progress in Behavior Modification, Vol. 7. Academic Press,
New York.
Rehm, L. P., Kornblith, S. J., O'Hara, M. W., Lamparski, D. M., Romano, J. M.,
and Volin, J. (1981). An evaluation of major components in a self-control
behavior therapy program for depression. Behavior Modification, 5, 459-490.
Rehm, L. P., Lamparski, D. M., Romano, J. M., and O'Hara, M. W. (in preparation).
Cognitive, behavioral and combined versions of a self-control therapy
program for depression.
Rogers, P. A., Kerns, R., Rehm, L. P., Hendlers, E. D. and Harkness, L. (1982,
August). Depression mitigation in hemo-dialysis: A function of self-
control training. Paper presented at the meeting of the American
Psychological Association, Washington, DC.
 98 L. P. Rehm
Romano, J. M., and Rehm, L. P. (1979, April). Self-control treatment of depres-
SiOn: One-year follow-up. In A. T. Beck (Chair), Factors affecting the
outcome and maintenance of cognitive therapy. Symposium presented at the
meeting of the Eastern Psychological Association, Philadelphia.
Rose&awn, M. (1978, November). A schedule for assessing self-control behavior:
Preliminary findings. Paper presented at the meeting of the Association
of Behavior Therapy, Chicago.
Roth, D., Bielski, R., Jones, M., Parker, W., and Osborn, G. (1982). A com-
parison of self-control therapy and combined self-control therapy and anti-
depressant medication in the treatment of depression. Behavior Therapy,
_12, 133-144.
Rothblum, E., Green, L., and Collins, R. L. (1979, April). A comparison of
self-control and therapist control in the treatment of depression. Pre-
sented at the meeting of the Eastern Psychological Association,
Philadelphia.
Seligman, M. E. P. (1975). Helplessness: On Depression, Development, and
Death. Freeman and Company, San Francisco.
Shaw, B. F. (1977). Comparison of cognitive therapy and behavior therapy in
the treatment of depression. Journal of Consulting and Clinical Psychology,
45, 543-551.
Spitzer, R. L., Gibbon, M. and Endicott, J. (1973). Global Assessment Scale.
New York State Department of Mental Hygiene, New York.
Spitzer, R. L., Endicott, J., and Robins, E. (1978). Research Diagnostic
Criteria: Rationale and Reliability. Archives of General Psychiatry,
2, 773-782.
Tharp, R. G., Watson, D. L., and Kaya, J. (1974). Self-modification of depres-
sion. Journal of Consulting and Clinical Psychology, 62, 624.
Thoresen, C. E., and Mahoney, M. J. (1974). Behavioral Self-Control. Holt,
Rinehart and Winston, New York.
Todd, F. J. (1972). Coverant control of self-evaluative responses in the
treatment of depression: A new use for an old principle. Behavior Therapy,
3, 91-94.
Tressler, D. P., and Tucker, R. D. (1980, November). The comparative effects
of self-evaluation and self-reinforcement training in the treatment of
depression. Paper presented at the meeting of the Association for the
advancement of Behavior Therapy, New York.
Vasta, R. (1976). Coverant control of self-evaluations through tempcral
cueing. Journal of Behavior Therapy and Experimental Psychiatry, 1, 35-37.
Wanderer, Z. W. (1972). Existential depression treated by desensitization of
phobias: Strategies and transcript, Journal
_- of Behavioral Therapy and
Experimental Psychiatry, 3, 111-116.
Wolpe, J. (1971). Neurotic depression: An experimental analog, clinical
syndromes, and treatment, American Journal of Psychotherapy, 25, 362-368.
Wolpe, J., and Lazarus, A. A. (1967). Behavior Therapy Techniques. Pergamon
Press, Oxford.