IMPROVEMENT IN CLIENTS WHO HAVE GIVEN DIFFERENT

REASONS FOR DROPPING OUT O F TREATMENT’

GENE PEKARIK*
LakelSumrer Community Mental Health Center
Leesburg. Florida

Addressed the dropout problem by (a) directly asking clients their reasons
for dropping out; and (b) assessing pretherapy to follow-up symptom change
in clients grouped by dropout reason. Forty-six dropouts who had been ad-
ministered the Brief Symptom Inventory at intake were contacted 3 months
later. At follow-up they were asked their reasons for dropping out and read-
ministered the Brief Symptom Inventory. It was found that 39% quit due to
“no need for services,” 35% due to “environmental constraints,” and 26%
due to “dislike of services.” The “no need for services” and “environmental
constraints” groups had significant decreases in intake to follow-up Brief
Symptom Inventory scores. Dropouts were found to be a heterogeneous
group whose follow-up adjustment was related to termination reason. This
contradicts prevailing notions that consider all dropouts as treatment
failures.
Literature reviews consistently have found that a large percentage, perhaps the ma-
jority, of psychotherapy outpatients terminate treatment by dropping out rather than by
mutual agreement with their therapists (Baekeland & Lundwall, 1975; Garfield, 1978).
This problem is especially acute in community mental health centers, where about 40%
of clients terminate after only one or two visits (Ciarlo, 1979; Fiester, Mahrer, Giambra,
& Ormiston, 1974; NIMH, 1979).
Dropouts usually are considered treatment failures, although almost no empirical
research has addressed this issue (Garfield, 1978). Research that has beem done on
dropouts has focused primarily on attempts to identify demographic and personality
characteristics of dropouts and their therapists (Baekeland & Lundwall, 1975). The
relationship between such variables and dropping out is modest at best, and findings are
frequently inconsistent (Baekeland & Lundwall, 1975; Garfield, 1978).
Perhaps due to the difficulty of contacting clients who have rejected services, little
effort has been devoted to addressing the dropout problem directly by asking clients their
reasons for termination. The two studies that have asked clients their reasons for drop-
ping out (Acosta, 1980; Garfield, 1963) found that environmental constraints, dis-
satisfaction with services, and perceived improvement were the most common reasons
given.
No information is available on the level of adjustment or degree of improvement at-
tained by clients who have different reasons for dropping out. Such information would be
particularly useful since most of the termination reasons given in the Acosta and Garfield
studies were clearly relevant to therapeutic outcome (e.g., perceived improvement, dis-
satisfaction with services).
The purposes of the present study were to (a) discover dropouts’ reasons for ter-
minating therapy; and (b) assess pretherapy to posttreatment follow-up changes in psy-
chiatric symptoms for groups of clients who have given different reasons for dropping
out.
‘Appreciation IS expressed to all the staff of Lake/Sumter Community Mental Health Center who
cooperated on this project.
T h e author is now at Washburn University. Reprint requests should be sent to Gene Pekarik, Ph.D.,
Department of Psychology, Washburn University, Topeka, Kansas 66621.
909
910 Journal of Clinical Psychology, November, 1983, Vol. 39, No. 6

METHOD
CIients
Clients were psychotherapy outpatients at a comprehensive community mental
health center that served small urban and semi-rural populations. From a group of 103
clients who had been classified as dropouts, 46 (45%) were contacted at follow-up.
Thirty-eight were contacted by phone and 8 by mail. Those contacted did not differ
significantly from those not reached on the following characteristics: Education, race,
sex, age, financial status as indicated by Title XX eligibility, and initial Brief Symptom
Inventory (Derogatis, 1977) severity level. The majority of clients were classified by
DSM I1 as Neuroses, Marital Maladjustment, Transient Situational Disturbances, and
Alcoholism.
Therapists
The 12 therapists who were treating the clients in this study described themselves as
eclectic in their therapeutic orientation. There were 3 Ph.D. psychologists, 6 with M.A.s
in psychology, 1 M.S.W., 1 B.S.W., and 1 with a B.A. in human services. Six therapists
were male, 6 female. They had an average of 59’2 years of professional experience.

Procedure and Measures

Pretherapy. Upon arrival at their intake appointments, clients were given a form
that explained the purpose of the research project, what their participation would entail,
reassurances of confidentiality, and the voluntariness of participation. The purpose of the
study was described as an effort to gain information about (a) the problems experienced
by clients; (b) clients’ reasons for deciding to continue or terminate services; and (c) ways
to improve the quality of services at the mental health center. It was emphasized that the
evaluation team was administratively independent of the clinics and would not report in-
dividuals’ follow-up information to the clinic staff. Those who agreed to participate (92%
of all admissions) then were asked to sign an Informed Consent Form and were given the
Brief Symptom Inventory (BSI), a 53-item self-report form that assesses the presence
and severity of a wide range of psychiatric symptoms (Derogatis, 1977).
The BSI is a shortened version of the Symptom Checklist-90 (Derogatis, 1977), with
which it correlates very highly; BSI-SCL-90 symptom dimension correlations range
from .92 to .99. Outpatient and nonpatient score norms are available for the BSI, which
takes about 10 minutes to complete. The BSI scores reported in this paper were obtained
by summing the distress level indicated for each of the 53 items.
Receptionists had been informed of the purpose of the study and were available to
answer clients’ questions about the study or the BSI. Standard demographic information
also was gathered at intake.

Follow-up. Three months after the intake date, follow-up contacts were attempted.
Clients with a telephone were called until contacted; those who did not answer their
phone were called a minimum of 10 calls, at systematic day and evening hours, over a 2-
week period. If still not contacted after these efforts, a questionnaire was mailed to the
client. If a reply to the mailed questionnaire was not received within 10 days, a reminder
letter was sent to the client. Clients without phones were sent a follow-up questionnaire
and a reminder letter if needed. All letters included stamped, addressed return envelopes.
Information was gathered in several areas at follow-up. The major procedure con-
sisted of: (a) a narrative readministration of the BSI (Derogatis [I9771 reports that no
bias is associated with narrative vs. written administration of the longer version of the
BSI, the Symptom Checklist); and (b), questions related to reasons for termination of
treatment. The same questions were asked on the mail-out questionnaire and in the
phone interview.
 Reasons for Dropping Out 91 1

Other client information. At each client’s termination, center therapists were re-
quired to assess the appropriateness of termination as part of standard clinic procedure.
This information was submitted routinely to the center’s computerized Management In-
formation System. By monitoring the center’s monthly Management Information
System printout, omissions of these termination assessments were detected and then ob-
tained from therapists. “Dropout” was defined as a client who was “in need of continued
treatment beyond his last session,” while an “appropriate termination” was defined as
someone “not in need of continued treatment beyond his last session.”
Client demographics and visit activity also were obtained through the center’s
Management Information System.
RESULTS
Reasons for Termination
Table I shows the various reasons given for dropping out of treatment and the
number and percentage of clients who gave each of the reasons. The largest percentage of
clients, 37’70, cited “no need for services” as their reason for terminating. The second
most popular reason for termination was “dislike of services,” endorsed by 17% of the
dropouts. No other single reason was given by more than 10% of those contacted.

TABLt I
Dropours’ Reasons for Therapy Termination

Reason N J,4

I . No need for services 17 37

7 Dislike of services 8 17
3 Preferred tredtment elseu here 4 9
4 TranSpOrtdtlOn problem\ 4 9
5 Conflict with work hour\ 3 7
6 FlndnCldl 2 4
7 Lack of time 2 4
x Moved 2 4
9. Program dropped by clinic I 2
10. Needed 10face problems on her oun I 2

I I. Hospitalized for medical problems I 2

I? Left town for several weeks I 2
-

All of the termination reasons fell into three broad categories: “No need for ser-
vices,” “dislike of services,” and “environmental constraints.” These are shown in Table
2, along with the termination reasons from Table 1, which were included in these broader
categories and the number and percentage of all dropouts included in each category.
With this grouping, the greatest percentage of clients (39%) were classified as citing “no
need for services,” 35% were classified as having had “environmental constraints which
interfered with treatment,” and 26% were classified as having had “dislike of services.”
Pretherapy to Follow-up BSI Changes for Dropout Subgroups
Table 3 shows the pretherapy and follow-up BSI scores for 41 of the 46 clients in the
dropout groups, along with the mean number of visits attended by each dropout group.
One-tailed r-tests for correlated pairs were performed to assess pre- to posttherapy
912 Journal of Clinical Psychology. November, 1983, Vol. 39, No. 6

changes for each of the three dropout groups. The “no need for services” and “en-
vironmental constraints” groups had significant decreases in BSI scores, while the
“dislike of services” group did not change significantly. There were no significant
differences among the groups for number of visits attended.

TABLE 2
Summary Categorization of Termination Reasonr

Category Termination reason (from Table I ) N 5%

No n d for services I. I I 18 39
Environmental constraints 4,5,6.7,8,9,10,12 16 35
Dislike of services 293 I2 26

TABLE 3
Mean BSI Scores and Numbers of Visits for Termination Groups
~~~~~ ~~~~~~ ~ ~~~ ~~

Pretherapy FOIIOW-UP
Termination category N Visits BSI BSI 1

No n d for services 16 2.3 56.4 38.6 2.79*

Environmental constraints 14 3.1 72.8 40.2 2.26*+
Dislike of services II 2.8 56.2 47. I .8 I

‘p <.01.
* p <.025.

DISCUSSION
The reasons given for c.opping out are inconsistent w.... the common belief that
dropouts are treatment failures. Like other studies of outpatient dropouts (Acosta, 1980;
Garfield, 1963). it was found that only a minority of clients said they dropped out due to
a dislike of services. That the most common reason given for dropping out was “no
further need for services” contradicts the dropout as failure notion. The percentage who
said that they had no need of services was higher than that found by Garfield or Acosta.
These differences may be attributable to Garfield’s extremely small sample size (N = 12)
and to Acosta’s high percentage of minority clients (68% of his sample).
It’s possible that clients in this and the other dropout studies were reluctant to
directly express negative feelings about the services they received. This could account for
the large percentage of clients in this, Acosta’s, and Garfield’s studies who cited en-
vironmental constraints as their reasons for dropping out. An inability to overcome such
obstacles as transportation problems and conflicting work hours might indicate a lack of
motivation for these clients.
The data on BSI changes for the dropout subgroups presented in Table 3 can be
used to address the veracity of clients’ dropout reasons. The lack of significant pretreat-
ment to follow-up symptom changes for the dislike of services group is what one would
expect from dissatisfied clients. Their dislike of services may well have stemmed from a
lack of symptom reduction. Similarly, the significant changes achieved by the “no need
for services” group is consistent with their stated termination reason and may have
formed the basis for their giving that reason. The interpretation of the “environmental-
 Reasons for Dropping Out 913

constraints” group’s significant reductions is less straightforward. It is plausible that

clients in this group allowed environmental constraints to disrupt treatment because
problem severity was reduced.
This study indicates that dropouts are not a homogeneous group of clinical failures.
A large percentage of dropouts consider themselves to be without need of treatment and
have achieved significant reductions in their symptoms over the course of their treatment.
REFERENCES
ACOSTA,F. X . Self-described reasons for premature termination of psychotherapy by Mexican American,
Black American, and Anglo-American patients. Psychological Reports. 1980. 47, 435-443.
BlreKeLmD, F., & LUNDWALL, L. Dropping out of treatment: A critical review. Psychologicol Bulletin.
1975, 82. 738-783.
CIARLO, J. A. Annual evaluation report for 1975 of the Northwest Denver Mental Health Center. In C. Win-
dle (Ed.),Reporting program evaluations: Two sample community mental health center onnual reports.
Rockville, Md.: Department of Health. Education and Welfare, 1979.
Deroorms, L. R. SCL-90: Adminisrrarion. scoring and procedures manual-I for the revised version.
Baltimore: Johns Hopkins University School of Medicine, 1977.
FIesTeR, A. R.. MAHRER. A. R., GIAMBRA. L. M.,& ORMISTON. D. W. Shaping a clinic population: The
dropout problem reconsidered. Community Mental Health JournaI. 1974, 10, 173- 179.
GARFIELD. S. L. A note on patients’ reasons for terminating therapy. Psychological Rcporrs. 1963. 13. 38.
GARFIeLD. S.L. Research on client variables in psychotherapy. In S. L. Garfield & A. E. krgin (Eds.).
Hand6ook ofpsychotherapy and behavior change (2nd 4 . ) . New York: John Wiley, 1978.
NATIONAL INSTITUTE OF MENTAL HeALTH. Report of the work group on health insurance, November 1974.
Cited in Ciarlo, J. A.. Annual evaluation report for 1975 of the Northwest Denver Mental Health Center.
In C. Windle (Ed.),Reporting Progrom Evoluationr: TwoSample Communiry Mental Healrh Center An-
nuol Reporrs. Rockville, Md.: Department of Health, Education and Welfare, 1979.