You are on page 1of 5

Cognitive-Behavioral Group Therapy With

Medication for Depressed Gay Men With


AIDS or Symptomatic HIV Infection
Mary R. Lee, M.D.
Larry Cohen, M.S.W.
Suzanne W. Hadley, Ph.D.
Frederick K. Goodwin, M.D.

Objective: The feasibility and effectiveness of a combination of group were found to predict a more rapid
cognitive-behavioral therapy and medication for the treatment of de- decline in CD4 lymphocyte counts
pression among gay men with AIDS or symptomatic HIV infection were (5). In another study, depressive af-
evaluated. Methods: Fifteen patients diagnosed with DSM-IV major de- fect was reportedly associated with
pressive disorder or dysthymia were treated in one of two weekly ther- increased mortality risk (6). These
apy groups in which cognitive-behavioral therapy had been specially findings indicate a need for effective
modified for the target population. The majority of these patients, in- treatments for depression for patients
cluding two who had been on medication before joining the groups, also with AIDS and HIV infection, partic-
received antidepressant medication. Thirteen of the 15 patients com- ularly those with more severe HIV
pleted therapy, attending an average of 15 of the 20 therapy sessions. disease.
Results: The group cognitive-behavioral therapy used in this project ap- Group cognitive-behavioral thera-
peared to be attractive to most patients; retention, attendance, and py, supportive group psychotherapy,
therapy compliance were good. Depression scores showed substantial and individual psychotherapies all
decreases from pre- to posttherapy, with further decreases at one-year have been reported to be effective in
follow-up. Patients’ self-reports indicated that some aspects of the in- reducing symptoms of depression or
tervention, particularly the focus on cognitive restructuring, were espe- distress among HIV-positive patients
cially valuable in alleviating their depression. Conclusions: The modi- (7–12). Group cognitive-behavioral
fied group cognitive-behavioral therapy described in this study report therapy was chosen for this project
offers a reasonable option for treatment of this clinically challenging because it combines several features
group of patients. (Psychiatric Services 50:948–952, 1999) shown to be beneficial in this popula-
tion. First, cognitive-behavioral ther-
apy teaches coping strategies to re-

S
everal studies have pointed to measures of depression (1,2). A “dra- duce stress and alter thoughts that re-
an increased risk for depression matic rise” in depressive symptoms inforce depression. Effective coping
among patients with AIDS or during the 18 months preceding an strategies have been shown to con-
symptomatic HIV infection. These AIDS diagnosis has been reported tribute to lower levels of depression
patients reportedly have a higher (3). Another study reported an in- among HIV-positive patients (1).
prevalence of depression than per- creased risk of suicide among persons Second, provision of social support
sons who are HIV negative or who are with AIDS (4). through the group may be especially
HIV positive but asymptomatic (1). Depression is harmful to physical important for patients with AIDS or
Likewise, HIV symptomatology re- health as well. In one study, overall symptomatic HIV infection who are
portedly is associated with elevated depression and “affective depression” often isolated from family members.
Satisfaction with support systems has
been shown to be inversely related to
depression (2). Third, the cognitive
Dr. Lee is affiliated with Clinical Studies, Ltd., in Falls Church, Virginia. Mr. Cohen is
a social worker in private practice in Washington, D.C. Dr. Hadley is senior scholar and therapy approach in this project— a
Dr. Goodwin is director of the program on medical science and society at the Ethics and structured, time-limited, goal-direct-
Public Policy Center, 1015 15th Street, Suite 900, Washington, D.C. 20005 (e-mail, ed intervention focused on current
sh845@eppc.org). Dr. Goodwin is also professor in the department of psychiatry and be- problems— is a treatment amenable
havioral sciences at George Washington University in Washington, D.C. Address corre- for use in managed care settings as
spondence to Dr. Hadley. well as community mental health cen-
948 PSYCHIATRIC SERVICES ♦ July 1999 Vol. 50 No. 7
ters, where many patients with AIDS sessment Scale (GAS) (17), and the Table 1
or symptomatic HIV infection receive Beck Depression Inventory (BDI)
Characteristics of 15 gay men with
care. (18)— were used throughout the ob-
AIDS or symptomatic HIV infection
This demonstration project was servation period.
at intake into cognitive-behavioral
conducted to evaluate the feasibility
group therapy with adjuctive medica-
and efficacy of group cognitive-be- Group cognitive-behavioral therapy
tion for treatment of depression
havioral therapy, described by Beck The cognitive-behavioral therapy was
and colleagues (13,14), plus adjunc- conducted by a licensed clinical social Characteristic N
tive antidepressant medication for worker specializing in private-prac-
treating depression among gay men tice cognitive-behavioral therapy in Age (years)
Mean 39
with AIDS or symptomatic HIV in- the gay community. Two treatment Range 26 to 54
fection. groups were conducted, the first with Race
nine patients and the second with six. Caucasion 12
Methods Some group cognitive-behavioral Black 3
Sample therapy procedures were adapted for College graduate 9
Living alone 8
Gay men with AIDS or symptomatic the project population. In particular,
Employed 7
HIV infection were recruited for the the treatment duration was extended History of depression 13
study through advertisements in a lo- to 20 weekly two-hour sessions, to History of substance abuse 4
cal paper serving the gay community permit more intensive work on dys- DSM-IV diagnosis
in Washington, D.C. Each prospec- functional core beliefs and help pre- Major depressive disorder 12
Dysthymia 3
tive enrollee underwent two clinical vent relapse. Patients were permitted
Out of bed more than
interviews, one by the project psychi- to continue any ongoing individual or half of day 15
atrist (the first author) and one by the support group therapy and to start Taking antidepressants at intake 2
project therapist (the second author). other individual or couples therapy, as Years since diagnosis of
Subsequently, the psychiatrist and needed. HIV infection
Mean 7
therapist derived a consensus DSM- The fundamental premise of the
Range 1 to 13
IV diagnosis. cognitive-behavioral therapy in this Type of HIV disease
To be included in the study, pa- project was that teaching patients to AIDS 12
tients had to meet criteria for major recognize and challenge irrational Symptomatic HIV infection 3
depressive disorder or dysthymia perceptions and beliefs will enable T cell count within six months
before intake
without active alcohol abuse or other them to interrupt depressive thought
Mean 158
drug abuse or dependence, acute sui- processes and relieve or prevent de- Range 32 to 450
cidality, bipolar disorder, borderline pressive symptoms. Several strategies
personality disorder, or psychotic dis- specific to cognitive-behavioral thera-
order. Eligible patients were offered a py were used. Specifically, automatic
course of cognitive-behavioral thera- thoughts were examined to illuminate sociated with health or relationship
py and, if indicated, adjunctive anti- their effect on mood and show how problems was common. During each
depressant medication. changing perceptions can avert de- session, the therapy goals served as an
Fifteen patients— 12 with major de- pressive thoughts and symptoms. In organizing point around which pa-
pressive disorder and three with dys- addition, dysfunctional core beliefs tients addressed problems and found
thymia— were enrolled after provid- about self-worth and expectations of solutions. The goals also were the ba-
ing informed consent. Twelve patients others were examined to demonstrate sis for self-help experiments, de-
had AIDS (category C of the criteria how they contribute to vulnerability signed in the group sessions and car-
of the Centers for Disease Control to depression. Specific instructions ried out in the intervening weeks.
and Prevention [CDC]) and three had were used to modify dysfunctional The basic format for the cognitive-
symptomatic HIV infection (category core beliefs, and peer modeling and behavioral therapy sessions was as fol-
B of the CDC criteria (15). Table 1 observational learning were used to lows. At the beginning of each ses-
shows other patient characteristics. enhance the therapy process. sion, patients completed the BDI,
The observation period for each pa- At the beginning of the group ther- and the results were reviewed by the
tient was 11 months, which included apy, each patient specified his therapy therapist, with particular attention to
five months of group cognitive-be- goals. Besides relief from depression, suicidality. Each patient reported on
havioral therapy and a six-month fol- patients’goals also focused on work— the previous week’s moods, thoughts,
low-up period. During the follow-up job loss was associated with increased self-help experiments, and home-
period, the project therapist and psy- depression and lowered self-worth— work. Worksheets and instructions for
chiatrist met with each patient at in- and anxiety about family relations. self-help experiments focusing on
tervals of six to eight weeks or as Family relations generally were poor, skills or concepts of current interest
needed. Objective measures of de- largely due to prejudice about the pa- were distributed.
pression and functioning— including tients’ sexual orientation. Patients’ This structured agenda occasionally
the Hamilton Depression Rating goals also included issues of relation- had to be modified to give additional
Scale (HDRS) (16), the Global As- ships or health. Substantial anxiety as- attention to a specific concern or cri-
PSYCHIATRIC SERVICES ♦ July 1999 Vol. 50 No. 7 949
sis. Under these circumstances, the recommended by the psychiatrist if Results
therapist would focus on the patient’s the patient had a sustained poor re- Although the focus of this descriptive
particular difficulty, asking other sponse to the group therapy or if the report is the adaptation of group cog-
group members to support and assist symptoms of depression significantly nitive-behavioral therapy and its fea-
in the process. This tactic promoted interfered with functioning. Two pa- sibility for use with gay men with
empathy among the other group tients were taking an antidepresssant AIDS or symptomatic HIV infection
members and enabled them to collec- at enrollment; otherwise, patients who are depressed, the patients’ re-
tively learn the skills taught to indi- generally were reluctant to start anti- sponse bears some mention. The cog-
vidual patients. depressants, citing prior nonre- nitive-behavioral therapy had a high
During the first ten sessions, the sponse, concerns about taking more retention rate: 13 of the 15 patients
relationship between depression and medication, and fear of side effects. completed the course of therapy with
automatic thoughts was examined us- The frequency of severe side effects good attendance (mean=15 sessions,
ing educational materials and the as- from antidepressant medications, with a range from eight to 20 ses-
signment of after-group exercises. Ex- particularly among patients with sions) and a high rate of compliance
ercises included the completion of AIDS, has been reported to be as with the follow-up assessment. One
weekly worksheets about specific life high as 25 percent (19). The rate of of the 13 died shortly after the end of
situations to practice the skills of cog- dropout due to side effects within six therapy and thus did not complete
nitive restructuring— identifying de- weeks of initiating medication has the outcome battery. Thus data were
pressive triggers and associated auto- been reported to be as high as 10 per- available for 12 patients at the end of
matic thoughts, finding the cognitive cent (20). therapy. Data were available for all of
distortions in those thoughts, and de- Despite the initial reluctance, all of those patients at follow-up.
veloping realistic and productive re- the seven patients offered antidepres- Outcome measures showed sub-
sponses. In addition, work was done sants during the project eventually stantial reductions in symptoms of
on a behavioral level to illuminate the accepted medication. Thus a total of depression, with the mean HDRS
relationship between mood and activ- nine of 15 patients in the project re- score declining from 26 at intake
ity. Because of this work, patients ceived cognitive-behavioral therapy (with a range from 8 to 38) to 9 after
were able early on to participate in plus medication. The shift in attitude therapy (a range from 1 to 18), and 6
mood-improving pleasurable activi- among these patients, which facilitat- at the follow-up (a range from 0 to
ties, with associated boosts in motiva- ed their acceptance of antidepressant 13). An improvement of more than
tion and morale. medications, may have resulted part- 50 percent in HDRS scores was ex-
In the second ten weeks of therapy, ly from the group therapy, where in- perienced by eight of 12 patients
relationships among dysfunctional formation about medication was pro- from baseline to end of therapy and
core beliefs, automatic thoughts, de- vided, fears about medications were by all 12 patients from baseline to fol-
pressed mood, and self-defeating be- discussed, and possible benefits of low-up.
haviors were examined. Materials medications were explained. A variety The mean BDI showed a parallel
used in these sessions included hand- of medications, including nortripty- decline, from 24 at baseline (with a
outs about the relationship between line, paroxetine, and bupropion, were range from 8 to 38) to 15 at end of
core beliefs and depression, question- used, depending on symptomatology therapy (a range from 2 to 39) to 9 at
naires to identify patients’ dysfunc- and on history of medication use and follow-up (a range from 0 to 23). The
tional core beliefs, and worksheets for side effects. BDI scores of four of 12 patients im-
detailed examination of core beliefs, All patients who accepted antide- proved by 50 percent or more from
including how they originate, their pressant medication responded with baseline to the end of therapy, and
advantages and disadvantages, and decreased symptoms and improved scores of seven patients improved by
how to generate new core beliefs. functioning. On average they re- 50 percent or more from baseline to
(Sample therapy worksheets are avail- mained on medication more than sev- follow-up.
able from the second author.) en months (a range from one month GAS scores increased from 57 at
to more than the 11 months of the baseline (a range from 40 to 70) to 72
Antidepressant medication study). Seven patients ultimately dis- at the end of therapy (a range from 60
All psychopharmacological treatment continued medication, two because to 90) to 80 at follow-up (a range from
was managed by the project psychia- they felt they were so improved that 65 to 90).
trist, who met with each patient at six- they no longer needed it and five be- The limitations of this study must
to eight-week intervals to evaluate the cause of side effects, including exces- be considered in interpreting these
need for antidepressant medication sive sedation, liver function abnor- results. The study was not controlled:
and, if indicated, to initiate or manage malities, and gastrointestinal upset. the small sample may not be broadly
medication. The psychiatrist, with the Four of these five patients remained representative, and the patients were
patient’s consent, had regular com- on medication at least two months permitted to use other therapies in
munication with the patient’s medical (mean=ten months), a period well in addition to the cognitive-behavioral
doctor about symptoms and antide- excess of the periods of six weeks or therapy. Thus the possibility that the
pressant medications. less commonly used to define med- patients’response was nonspecific or
Antidepressant medication was ication dropout (20). due to factors other than cognitive-
950 PSYCHIATRIC SERVICES ♦ July 1999 Vol. 50 No. 7
behavioral therapy cannot be ruled pressed HIV-positive subjects. The AIDS may be a vital resource for this
out. Eleven patients used another study also showed that improved cog- population. The most seriously physi-
therapy concurrent with cognitive- nitive coping skills were strongly re- cally ill patients in our project seemed
behavioral therapy: two used psy- lated to decreased dysphoria, anxiety, at greatest risk for a relapse into de-
chotherapy, five used medication, and and measures of total disturbance and pression. And, as previously noted,
four used psychotherapy plus med- mood disturbance. hopelessness and depression, in turn,
ication. However, several patients had may actually exacerbate physical de-
a history of nonresponse to such Discussion cline (5,6). For all of these reasons,
treatments. More important, several Among the patients in this project, treatment-resistant patients may very
patients made comments about fac- the grim reality of their illness was a well be the ones most in need of the
tors specific to cognitive-behavioral significant chronic stressor, which remedial and preventive effects of
therapy that they believed had con- seemed to lower the patients’thresh- cognitive-behavioral therapy.
tributed to their improvement. old for depressive responses to acute In contrast to the findings of this
At follow-up, each patient was triggers. In other words, because of project, some studies have failed to
asked to describe the aspects of ther- their physical illness, patients ap- show an advantage of cognitive-be-
apy most helpful to him. Several peared chronically hypervulnerable havioral therapy over other treatment
themes emerged in these self-reports. to hopelessness, helplessness, and modalities for patients with AIDS or
First, the therapy group provided so- self-blame associated with a variety of symptomatic HIV infection. Among
cial support for patients who previ- stimuli. These circumstances created the reasons for these findings may be
ously felt isolated. Second, by observ- obstacles to the core work of the cog- the low pretreatment depression
ing comrades, patients learned that scores in some studies, which may
their problems were experienced by have created a “floor effect” and left
others, which decreased feelings of little room for improvement (8,9).
isolation and shame and allowed con- Another factor may be the short dura-
structive interchanges about solutions
Patients who tion of treatment in some studies,
or coping strategies. Third, the expe- such as eight sessions (7). Our experi-
rience of empathy for others’distress
are treatment resistant ence indicates that an adequate trial
contributed to a resurgence of self-es- of cognitive-behavioral therapy for
teem and was itself healing.
may very well be the ones depressed patients with AIDS or
In addition, several patients report- symptomatic HIV infection should be
ed that learning about cognitive re-
most in need of the remedial considerably longer. A remaining
structuring had enabled them to fo- question is whether a longer treat-
cus on positive aspects of life, to use
and preventive effects of ment period or a series of booster ses-
support and help from others more sions might be even more effective
effectively, to recognize triggers for
cognitive-behavioral for promoting cognitive restructuring
depression, and to develop internal and preventing relapse.
skills to modify depressive thoughts.
therapy. Concerning the therapy content,
Their comments included “I’m glad several themes emerged that under-
to have learned the tools. I feel I scored the complex interaction of bi-
know how to stop a lot of depression- ological, psychological, and social fac-
causing or -deepening situations,” nitive-behavioral therapy— the ex- tors associated with AIDS and symp-
“Group helped me train my mind to plicit and direct challenge of auto- tomatic HIV infection. Besides gen-
focus on good things,” “[I] learned matic thoughts and dysfunctional eralized guilt— “Look how I screwed
about distortions. I make sure I hear core beliefs— and made the cogni- up my life”— themes included guilt
what the [other] person is saying, not tive-behavioral therapy more arduous about contracting HIV, AIDS as a
just assuming,” “Am able to see how and prolonged. punishment for being gay, and guilt
my own thoughts influenced how I In addition, although the patients about the possibility of infecting oth-
felt, that there was a lot more internal in the study generally could use be- ers. Anger— toward the medical com-
control than I thought,” and “The havioral techniques to improve mood, munity, lovers, or the therapist— was
cognitive therapy approach was help- some resisted working on cognitive common and often seemed dispro-
ful. I would be dead without it.” distortions and core beliefs, based on portionate to the supposed stimulus.
Supportive data pointing to specific the attitude of “Why bother? I’m go- The appearance of new symptoms of
mechanisms of cognitive-behavioral ing to get sicker and die; it’s no use.” physical illness was often associated
therapy’s efficacy have recently been This resistence to cognitive restruc- with a relapse into depression, includ-
published. In a controlled study, Lut- turing was particularly unfortunate, ing cognitive distortions like “My life
gendorf and associates (11,12) found because, as noted by Lutgendorf and is no good anymore; there is no value
that cognitive-behavioral therapy associates (12), coping strategies that in life.” Within the group, one mem-
(plus relaxation training) was associat- facilitate the acceptance and integra- ber’s becoming more ill often trig-
ed with significant improvement in tion of life changes that are associated gered hopelessness and fear among
cognitive coping skills among nonde- with symptomatic HIV infection and others.
PSYCHIATRIC SERVICES ♦ July 1999 Vol. 50 No. 7 951
The patients also expressed a sense reports indicated that the interven- 11. Lutgendorf SK, Antoni MH, Ironson G, et
al: Cognitive-behavioral stress manage-
of having lost value to others— “I am tion’s emphasis on cognitive restruc- ment decreases dysphoric mood and her-
no longer attractive”— which inter- turing was particularly helpful in en- pes simplex virus–type 2 antibody titers in
fered with the formation of new rela- abling patients to cope with and pre- symptomatic HIV-seropositive gay men.
Journal of Consulting and Clinical Psychol-
tionships, especially with other HIV- vent the recurrence of depression. ogy 65:31–43, 1997
positive people: “What’s the use? We believe that the approach com-
They’re going to die too.” Last, group bining modified group cognitive-be- 12. Lutgendorf SK, Antoni MH, Ironson G, et
al: Changes in cognitive coping skills and
members frequently focused on phi- havioral therapy and adjunctive med- social support during cognitive-behavioral
losophy-of-living issues rarely men- ication that was used in this project stress management intervention and dis-
tioned in other depression treatment may be practical and useful for treat- tress outcomes in symptomatic human im-
munodeficiency virus (HIV)–seropositive
groups, such as “What do I want to do ing patients with AIDS or sympto- gay men. Psychosomatic Medicine 60:
with my life?” and “How is life to be matic HIV infection who are de- 204–216, 1998
lived, as long as possible or as fully as pressed, particularly in managed care 13. Rush AJ, Beck AT, Kovacs M et al: Com-
possible?” These questions frequent- and mental health center settings, parative efficacy of cognitive therapy and
ly took on imminent importance. It and is worthy of further study. ♦ pharmacotherapy in the treatment of de-
pressed outpatients. Cognitive Therapy Re-
was often the sickest members who search 1:17–37, 1977
posed these questions; in so doing, Acknowledgment
they allowed others to talk about their This project was supported by a bequest 14. Shaw BG: A comparison of cognitive thera-
from the estate of Paul Garofalo of San py and behavior therapy in the treatment of
own fears of the future. depression. Journal of Consulting and Clin-
Francisco.
Concerns about dying were com- ical Psychology 45:543–551, 1977
mon, especially in the first therapy References 15. Centers for Disease Control: 1993 revised
group. One patient deteriorated dur- classification system for HIV infection and
1. Folkman S, Chesney MA, Pollack L, et al:
ing the second half of the group’s ses- Stress, control, coping, and depressive
expanded surveillance case definition for
sions, dying shortly thereafter. The AIDS among adolescents and adults. Mor-
mood in HIV positive and negative gay men
bidity and Mortality Weekly Report 41(RR-
availability of protease inhibitors and in San Francisco. Journal of Nervous and
17):1–19, 1992
Mental Disease 181:409–416, 1993
the hope for increased survival be-
16. Hamilton MD: Diagnosis and rating of de-
came known during the later weeks of 2. Hays RB, Turner H, Coates T: Social sup-
pression. British Journal of Psychiatry 3:76–
the second therapy group. With this port, AIDS-related symptoms, and depres-
79, 1969
sion among gay men. Journal of Consulting
new possibility for treatment, pa- and Clinical Psychology 60:463–469, 1992 17. Endicott J, Spitzer RL, Fleiss JL, et al: A
tients’ underlying assumptions about procedure for measuring overall severity of
the inevitability of death shifted. The 3. Lyketsos CG, Hoover DR, Guccione M, et
psychiatric disturbance. Archives of Gener-
al: Changes in depressive symptoms as
resulting hopefulness seemed to al- al Psychiatry 33:766–771, 1976
AIDS develops. American Journal of Psy-
low these patients to focus more con- chiatry 153:1430–1437, 1996 18. Beck AT, Ward CH, Mendelson M, et al:
structively on their treatment. 4. Marzuk PM, Tierney H, Tardiff K, et al: In- An inventory for measuring depression.
The therapist’s qualities also were creased risk of suicide in persons with Archives of General Psychiatry 4:561–571,
AIDS. JAMA 259:1333–1337, 1988 1961
important to the treatment’s effects.
Patients’reports pointed to the thera- 5. Burack JH, Barrett KC, Stall RD, et al: De- 19. Hintz S, Kuck J, Perekin JJ, et al: Depres-
sion in the context of human immunodefi-
pist as a vital element in the group’s pressive symptoms and CD4 lymphocyte
ciency virus infection: implications for
success. During the cognitive-behav- decline among HIV-infected men. JAMA
270:2568–2573, 1993 treatment. Journal of Clinical Psychiatry
ioral therapy sessions, the therapist 51:497–501, 1990
was an active instructor, coach, and 6. Mayne TJ, Vittinghoff E, Chesney MA, et
20. Wagner GJ, Rabkin JG, Rabkin R: A com-
al: Depressive affect and survival among
facilitator. He was also available to pa- gay and bisexual men infected with HIV. parative analysis of standard and alternative
tients for crisis management between Archives of Internal Medicine 156: 2233– antidepressants in the treatment of human
immunodeficiency virus patients. Compre-
therapy sessions. Personal qualities 2238, 1996
hensive Psychiatry 37:402–408, 1996
were especially important, as reflect- 7. Kelly JA, Murphy DA, Bahr GR, et al: Out-
ed in these patients’ reports of what come of cognitive-behavioral and support
was most helpful: “The therapist was group brief therapies for depressed HIV-in-
fected persons. American Journal of Psy-
accepting and understanding and chiatry 150:1679–1686, 1993
willing to learn what it means to be
8. Markowitz JC, Klerman GL, Clougherty
HIV positive so that I felt he was gen- DF, et al: Individual psychotherapies for
uinely supportive and cared— he is depressed HIV-positive patients. American
the best I’ve seen or had the pleasure Journal of Psychiatry 152:1504–1509, 1995
of talking with,” “[He] is gentle and 9. Mulder CL, Emmelkamp MG, Antoni
non-confrontational,” and “I trusted MH, et al: Cognitive-behavioral and experi-
[him].” ential group psychotherapy for HIV-infect-
ed homosexual men: a comparative study.
Psychosomatic Medicine 56:423–431, 1994
Conclusions
10. Targ EF, Karasic DH, Diefenbach PN, et
The structured group cognitive-be-
al: Structured group therapy and fluoxetine
havioral therapy in this project was at- to treat depression in HIV-positive persons.
tractive to most patients. Follow-up Psychosomatics 35:132–137, 1994

952 PSYCHIATRIC SERVICES ♦ July 1999 Vol. 50 No. 7

You might also like