University of Southern California Dissertations and Theses

Improvement Rate In Tuberculosis As A Function Of Accessibility To Repressed Emotional Disturbance

(USC Thesis Other)

Improvement Rate In Tuberculosis As A Function Of Accessibility To Repressed Emotional Disturbance

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content IMPROVEMENT RATE IN TUBERCULOSIS
AS A FUNCTION OF ACCESSIBILITY
TO REPRESSED EMOTIONAL DISTURBANCE
by
H arry J . Rosenthal
A D issertation P resen ted to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In P a rtia l Fulfillment of the
Requirem ents for the Degree
Doctor of Philosophy
(Psychology)
January 19 59
UNIVERSITY OF SOUTHERN CALIFORNIA
GRADUATE SC HOOL
UNIVERSITY PARK
LO S A NGELES 7
This dissertation, written by
...................Uar.ry..J.,...RQJseja.thai......................
under the directioti of.hiS-Guidance Committee,
and approved by all its members, has been pre
sented to and accepted by the Faculty of the
Graduate School, in partial fulfillment of re
quirements for the degree of
DOCTOR OF PHILOSOPHY
Date Jamiar.y...lJ9£x9.
c k CBa n c e c o m m i t t e e
/ V Chairman
...........
To My Wife
Ann
TABLE O F CONTENTS
CH A PTER PAGE
A C K N O W L E D G M E N T S ................................................................................... iii
I. BACKGROUND O F THE S T U D Y ...................................................... 1
M edical and D em ographic A s p e c t s ............................................ 1
Psychological F a c t o r s ..................................................................... 4
The Psychological V a r i a b l e ............................................ 6
II. THE P R O B L E M ........................................................................................ 8
Survey of the L iteratu re on Psychological F a c to rs . . 8
Etiological r e la tio n s h ip s ........................................................... 9
Psychological factors and irre g u la r discharge . . . 11
Im provem ent r a t e .......................................................................... 13
Sum m ary of L i t e r a t u r e ..................................................................... 16
Rationale of P r e s e n t S t u d y ........................................................... 17
Statement of the P r o b l e m ................................................................ 23
H y p o t h e s e s .............................................................................. 24
III. M E T H O D ....................................................................................................... 25
S u b j e c t s ............................................................... 2 5
I n s t r u m e n t s ............................................................................................. 29
Affective rating s c a l e ................................................................ 31
P r o c e d u r e ............................................................................................. 32
Selection of neutral and emotionally loaded words . 33
Selection of emotionally loaded w o r d s .............................. 34
V
C H A PTER PAGE
Selection of neutral words . . . . . . . . . . . . 34
F re e association to emotionally loaded
and neu tral w o r d s ........................................................... . 34
M easu res and Anticipated R e s u l t s ................................... 35
A ccessibility to re p re s se d emotional disturbance
is m ea su re d in two w a y s ........................................ 35
Rate of physical i m p r o v e m e n t ................................... 36
IV. R E S U L T S .............................................................................................. 37
Statistical D e s c r i p t i o n ............................................................ 39
H y p o t h e s e s .................................................................................... 46
A ccessibility ratio (L/N+-L) in relation to
im provem ent t i m e ....................................................... 46
A ccessibility in te r m s of num ber of loaded words
rated zero in relation to im provem ent time . . . 49
Supplementary D a t a ................................................................. 51
Occupational s t a t u s ............................................................ 51
Loaded words irresp ectiv e of disturbance rating . . 51
Loaded words rated d i s t u r b i n g ................................... 54
V. DISCUSSION......................................................................................... 59
Suggested A reas for Future R e s e a r c h ......................... 64
VI. SUMMARY AND C O N C L U S IO N S ............................................. 67
M e t h o d .............................................................................................. 68
vi
C H A P T E R PAGE
R e s u l t s ........................................................................................................ 70
C o n c lu s io n s ............................................................................................... 71
A P P E N D I X ............................................................................................................. 72
R E F E R E N C E S ......................................................................................................... 75
LIST O F TABLES
TABLE PAGE
1. Education, Age, Occupational Status, and Ethnic Group
of the G re a te r and L e s s e r Pathology G r o u p s ................... 27
2. Significance of the Differences between Means of
Improvem ent Tim e M easured by Time from Beginning
of Hospitalization to Date of Reaching Sem i-am bulatory
Status (Class IV) for All G r o u p s ................................................ 41
3. Significance of the Differences between Means of
A ccessibility Scores M easured as the Ratio between
Num ber of words Associated to Loaded W ords Rated
Zero, and the Sum of W ords Associated to Both Loaded
and Neutral Words (L/N+L) for All G r o u p s ....................... 42
4. Significance of the Differences between Means of A ccessi
bility Scores M easured as the Number of Words Loaded
but Rated Zero in Disturbance by Subjects for All
Hospital G r o u p s .................................................................................. 43
5. Rate of Improvement in Relation to Both M easures of
Accessibility (L/N+L and N um ber of Loaded Words
Rated Zero in Disturbance) for Combined Hospital
G r o u p s ..................................................................................................... 48
v iii
T A B L E PA G E
6. Rate of Im p ro v em en t in R elation to Both M e a s u re s of
A ccessib ility (L/N-tL and N u m b er of Loaded W ords
Rated Z e ro in D isturbance) for Long B each VAH
L e s s e r Pathology G r o u p .................................................................. 50
7. O ccupational Status in Relation to Im p ro v em en t Rate
for L e s s e r and G r e a te r Pathology G r o u p s .......................... 52
8. O ccupational Status in Relation to Both M e a su re s of
A ccessib ility (L/N+L and N u m b er of Loaded W ords
R ated Z e ro in D isturbance) for L e s s e r and G r e a te r
Pathology G r o u p s ................................................................................. 53
9. Rate of Im p ro v em en t in R elation to Loaded W ords
Irre s p e c tiv e of D isturbance Rating, for H ospital
G roups Com bined on B a s is of S im ila r Pathology and/or
S im ila r M eans of E x p e rim e n ta l V a r i a b l e s .......................... 55
10. N u m b er of Loaded W ords Rated D isturbing in Relation
to Im p ro v em en t Rate for L e s s e r and G r e a te r D egree
of Pathology G r o u p s ............................................................................ 56
11. N u m b er of Loaded W ords Rated D isturbing in Relation
to Both M e a su re s of A c cessib ility for G r e a te r and
L e s s e r Pathology G r o u p s .................................................................. 57
LIST O F FIGURES
FIGURE PAGE
1. Frequency distribution of time elapsed from adm ission
date until reaching C lass IV for le s s e r and g re a te r
pathology groups ............................................................................... 40
2. Frequency distributions of accessibility ratio sco res
(L/N+L)of le s s e r and g re a te r pathology groups . . . . 45
3. Frequency distributions of accessibility sc o re s m easu red
as words rated “ not disturbing” but loaded according
to complex indicators for g re a te r and le s s e r
pathology g r o u p s .................................................................................... 47
CHAPTER I
Background of the Study*
The purpose of this study is to investigate a possible relatio n
ship between accessibility to re p re s se d emotional disturbance and
im provem ent rate in tuberculosis.
By way of introduction, brief com m ent will be made regarding
the m edico-clinical and dem ographic aspects of the d isea se. Next, the
general im portance of psychological factors in tuberculosis will be
pointed out. Lastly, a b rie f statem ent will be made of the reaso n s for
initially selecting the variables of the study.
Medical and Dem ographic A spects
According to the National T uberculosis A ssociation (42) in 1952
there w ere 1,200,000 cases of tuberculosis in the United States,
400,000 of them active. At the present time despite the advances in
chem otherapy, it is thought that the incidence rate is approxim ately
the sam e.
The h isto rical aspect of the disease is te rse ly described by
*The author gratefully acknowledges the co-operation of M r.
E . V. E dw ards, M anager, Long Beach V eterans A dm inistration
Hospital, D r. F ran k K irkner, Chief Clinical Psychologist, and D r.
Kenneth Smith, Chief, P ulm onary D isease S ervice. Appreciation is
also ex p ressed to M r. F ra n c is W. Rollins, M anager, San Fernando
V eterans A dm inistration Hospital, and D r. B a rb a ra Stew art, Chief
P sychologist.
2
C ecil and Loeb as follows;
Archaeologic discoveries of skeletons bearing m ark s of tu b e r
culosis lesions indicate that the disease was present in rem ote
antiquity. In the e a rlie st m edical re c o rd s it was called consum p
tion or phthisis because of its m ost conspicuous feature, wasting.
Long considered to be of many varieties and origins, the unity of
the disease was first recognized by Laennec. The c o rre c tn e ss of
his conception was not universally acknowledged until Koch in 1882
isolated the specific organism and reproduced it experim entally.
The avian type of bacillus was isolated in 1890 by Mafucci, and the
bovine type in 1898 by Theobald Smith. (10:254)
According to Dubos and Dubos (25) the m ortality rate in 1950
was 22.2 p er 100,000 in the United States. However, they found that in
countries where a lower standard of living prevails there a re higher
m ortality ra te s. F o r exam ple, in Japan in that y e a r the rate was 240
p e r 100,000.
Tuberculosis can attack any part of the body, but pulm onary
tuberculosis is by far the com m onest form of the disease and it is this
form with which the p resen t study deals.
The d isease is communicable and infection takes place in the
m ajority of cases by breathing the bacilli, whether in dust o r droplets.
Two kinds of lesions a re produced: p rim a ry and reactivated. The
fo rm e r are residues of subclinical infection usually contracted in
childhood, which heals without observable sym ptom s. The reactivated
type re p re se n ts the reopening of previously healed lesions, and p r o
vides the basis of m ost adult form s of tuberculosis, although p rim a ry
infection is becoming m o re prevalent among adults.
3
Since “ rate of re c o v e ry ” is one of the variables of this study,
it is n e c e ssa ry to deal briefly with the m an n er in which p ro g ressio n of
healing takes place. The initial phase, after infection of lung tissue,
consists largely of a coagulative p ro cess in which dead tissue form s a
cheese-like ab sc e ss. If unchecked, the lesion will liquify and eventu
ally slough off to infect the healthy tissu e. This evacuation of necrotic
m a tte r which harbors the infection-producing bacilli leaves a cavity
rem aining in the lung. Healing o r re p a ir of the active lesion can occur
at any of the stages just outlined. The healing p ro cess is essentially
one of encapsulating the focus of infection, thus preventing spread of
bacilli. This is accom plished by fib ro sis, _i. e. growth of the fibrous
connective tissue which encases the lesion. Recovery (not cure) is
said to occur if the tubercle produced by fibrosis undergoes calcifi
cation (29),
Until recently, treatm ent methods w ere all indirect. The basic
treatm en t for tuberculosis is still bed re st. This im plies m ental and
emotional relaxation as well as physical re s t. Adequate diet is also a
necessity. Various su rgical procedures a re often used in o rd e r to
help the lungs re st and to facilitate reco v ery . These procedures
include collapsing the lung tem p o ra rily (pneumothorax, pneum operi
toneum) o r perm anently (thoracoplasty); operating on the phrenic nerve
so as to stop the m ovem ent of one side of the diaphragm for a num ber
of months (phrenic crush); and rem oving the infected parts of the lung
4
(re se c tio n ).
Drugs which directly attack the bacillus re p re se n t a great
stride in shortening the length of tre atm e n t and reducing the frequency
of the need for su rg e ry . The drugs m ost frequently used are s tre p to
mycin and p ara-am in o -salicy lic acid (PAS), and isoniazid (INH).
Occasionally, these drugs produce toxic “ side effects” and it has been
dem onstrated that tubercle bacilli can develop “ r e s is ta n c e ” to them .
At any rate, while the drugs are usually very effective, they cannot be
considered “ c u r e s ” in the light of present evidence.
It is generally held that a m ultiplicity of factors, e .g . sex,
ra c e , extent of lesion, socioeconomic statu s, etc., determ ine individ
ual response to tre atm e n t.
Psychological F acto rs
Wittkower (64,65) states that the emotional reactions of tu b e r
culosis patients have been a m a tte r of in te re st since the days of
H ippocrates. This in tere st has continued to the present, finding e x
pression not only in scientific writing but also in the productions of
novelists, poets, and opera lib rettists. As a result, m any som etim es
contradictory legends have a risen about the patient. F o r exam ple, he
is unduly cheerful (spes phtisica); he is depressed; he is a genius; he is
lacking in intelligence; he is likely to be psychotic; he has excessive
sexual appetite. It is only recently (32,64,56) that re se a rc h has tended
5
to show that th ere is no single o r predom inant type of p erso n who
develops tub ercu lo sis.
B a r k e r ’s review (2) of the lite ra tu re p rio r to 1953 points up the
confusion which c h a ra c te riz e d the professio n al thinking about tu b e r
culosis and its concom itant a n d /o r antecedent psychological factors up
to that tim e. Opinion runs from the one ex trem e of overlooking p s y
chological factors to the other, which alm o st ignores the contribution
of the tu bercle bacillus to the developm ent of the d ise a se . H ow ever, it
a p p ears that alm o st all physicians who have used the psychosom atic
approach in the study of tu berculosis have rep o rted that personality
factors play a m a jo r ro le . Thus it is that W ittkower (63:161) sta te s it
is “ so m etim es s a fe r to a s s e s s the patient’s prognosis on the b asis of
his p ersonality and of his em otional conflicts than on the basis of the
shadow on the film .”
The b earing of em otional factors on m etabolic functions has
also long been recognized. Cannon (9), Wolff (66), Selye (53), and
Mahl (40) am ong others se e m to have provided physiological links
between em otional factors and s tru c tu ra l change. H olm es (32) has
recently rep o rted on com prehensive studies which suggest that a d re n o
c o rtic a l activity plays a role in re sista n c e to tu berculosis and that the
effects of life s tr e s s upon the course of tu b ercu lo sis, in part, m ay be
m ediated via the ad renal gland.
In the last five y e a rs studies concerned with psychological
6
v a ria b le s in tu b e rc u lo sis have g reatly in c re a se d in n u m b er. H ow ever,
few of th ese studies d eal with the v ariab le of im p ro v e m e n t ra te and, in
the m ain, even these few a re vaguely fo rm ulated and inconclusive.
The P sy c h o lo g ical V ariable
The specific psychological v ariab le se le c te d for study as
possibly re la te d to im p ro v e m e n t ra te among tu b e rc u lo sis patients is
that of a c c e ssib ility to r e p r e s s e d a r e a s of em otional d istu rb a n c e . It
has re fe re n c e to the way in which patients v a ry in (a) the ability to free
a s so c ia te in such a r e a s , and (b) the n u m b er of th ese a r e a s . This
ac c e ssib ility has re fe re n c e to the c lo se n e ss, so to speak, of a person
to his conflicts.
The possible rele v a n c e of a c c e ssib ility to em otional conflict to
im p ro v e m e n t rate in ch ro n ic d ise a se firs t o c c u rre d to the w r ite r while
interning at Long Beach V eteran s A d m in istratio n H ospital. In the
c o u rse of psychotherapeutic and psychological sc re e n in g interview s
with tu b e rc u lo sis patients, he was im p re s s e d by the varying ability of
these patients to adequately v erb alize about a n d /o r to becom e fully
aw are of a r e a s of em o tio n al conflict. P r e lim in a r y inspection of the
hospital re c o rd s suggested a relatio n sh ip betw een im p ro v em en t ra te
and ability to talk about r e p r e s s e d em otional conflict. L a te r it was
noted for a s m a ll group of patients roughly hom ogeneous with r e g a rd to
extent of d ise a se on a d m issio n , so cio -eco n o m ic background, age, and
7
kind of m edical therapy that the sam e relationship seem ed to exist.
Search of the literatu re at that tim e brought to light only one
study having to do with the variables considered h ere. Colem an (16),
on the basis of psychoanalytic interview s with tuberculosis patients,
reported that effectively re p re s se d hostility and overtly expressed
hostility w ere directly related to im provem ent; suppressed hostility
was negatively related to im provem ent. However, his study did not
conform to s tric t experim ental standards: the variables w ere not
clearly defined and the subjects were not sufficiently sim ila r.
F u rth e rm o re , conflict about hostility is only one of the many
possible conflict a re a s which might be re p re s se d . Hence, it seem ed
m o re fruitful to investigate the general ability to verbalize to any
pertinent re p re sse d conflict a re a s and the num ber of such a re a s.
The need for such a study also seem s to be implied in the
following statem ent by Coburn:
In the diagnosis, treatm en t, and the rehabilitation of people
suffering from tuberculosis, em otional factors are of param ount
im portance. . . . We m ust have techniques to find out each patient’s
problem s and the flexibility to m eet these problem s once d is
covered. The m ajor technique psychiatrists have to offer those
responsible for the tuberculous is sim ple in w ords— Get the patient
to talk about h im se lf. Most people enjoy the opportunity but at the
sam e time may have a tendency to avoid the significantly d istu rb
ing a r e a s . (13:299)
C H A PT E R II
The P ro b lem
This chapter includes a su rv ey of the lite ra tu re relevant to the
variables involved in this study. Next, the rationale for the gen eral
hypotheses will be d iscu ssed . Following this, a statem en t of the p ro b
lem and the specific hypotheses will be m ade.
Survey of the L ite ra tu re on P sychological F a c to rs
Investigation rev eals that th ere are few studies relating to the
specific v a riab les involved in this study. However, under the b ro ad er
heading of psychological factors and tub ercu lo sis, the lite ra tu re is
extensive. S ev eral good reviews of the whole a r e a are available.
Among these m ay be m entioned those of B a rk e r (2), B e rle (5), M e rrill
(41), and M oorm an (44). P ossibly m o st significant both as a review of
the lite ra tu re and as a collection of new studies is the recen t publica
tion edited by S p a re r (55).
Since the lite ra tu re is extensive, only those studies bearing
closely on the v ariables of the p resen t one will be d iscu ssed . Such
lite ra tu re can be conveniently tre ated in te rm s of the relationship of
psychological factors to (a) etiology of tuberculosis, (b) irre g u la r
d isc h a rg e s, L £. d isch a rg e s obtained against m edical advice, the
patient leaving the hospital eith er with o r without the knowledge of
9
m ed ical personnel, and (c) im pro v em en t ra te .
E tiological Relationships
M ost authorities a re ag reed that tu berculosis is m ulticausal,
but that psychological factors a re im portant. Kaplan and Kaplan (3 6)
d e sc rib e it as a d is o rd e r w here psychological factors a re of definite
significance but secondary to an apparent p rim a ry organic p ro c e ss .
Day states:
To develop chronic active pulm onary tu b ercu lo sis a p erso n
needs som e bacilli, som e m od erately inflam m able lungs (not
celluloid like the guinea pig’s n o r asbestos like the elephant’s)
and some in tern al o r ex tern al factor which low ers the re sista n c e
to the d ise a se . . . . I am convinced that in so m e c a se s d isturbance
of the psyche can and does precipitate the active d isea se p ro c e ss.
(18:51)
H endricks s u m m a riz e s the c u rre n t im p re ssio n as follows:
It has not been sufficiently s tr e s s e d that there is m o re to the
etiology of tuberculosis than the p resen ce of tu bercle bacilli. The
two m o st pronounced facto rs a re u ndernourishm ent and fatigue.
P ra c tic a lly all the following neurotic tren d s form a basis for bad
living and eating habits, which, in turn, tend to u n d ernourishm ent
and fatigue. E m otional instability is the factor that drives the
patient to overw ork, ir r e g u la r hours, inadequate m eals, loss of
sleep, and over-indulgence. (31:208)
While su b scribing to the principle of m ulticausality, Dunbar
(26) nonetheless m aintains that pulm onary tu b ercu lo sis patients are
s im ila r in c h a ra c te r tra its , in past histo ry , and in m ode of life p rio r
to onset of d ise a se (diagnostic profile). She also sta te s that the high
incidence of neurotic tre n d s existed p rio r to the d ise a se and are
m e re ly accentuated by it. F o r “ neurotic tr e n d s ” she states one might
10
su b stitu te “ p a r tic u la r failu re of adaptation in a s t r e s s s itu a tio n .”
W ittkow er (63) in d icates th e r e is no diagnostic p ro file , but
o b s e rv e s that a dependency conflict is the distin g u ish in g c h a r a c te r is tic
of p e rso n s who develop tu b e rc u lo s is . F u r th e r m o r e , he s ta te s that an
inordinate n eed for affection is an outstanding fe a tu re of the p re m o rb id
p erso n a lity . T his need m ay be openly e x p re s s e d , thinly d isg u ised ,
w e ll-c o n c e a le d , o r flatly denied. In a n o th e r study involving 80 p atients,
W ittkow er, D u ro st, and L.aing (64) found five ty p es of tu b e rc u lo sis
patients which they c la ssifie d a s “ o v e rtly in s e c u re , s e lf - a s s e r t i v e ,
s e lf-d riv in g , re b e llio u s , and c o n f l i c t - h a r r a s s e d .” H o lm es (32)
e m p h a siz e s the im p o rta n c e of c u ltu ra l conflicts: ethnic, ra c ia l, and
econom ic m in o rity s ta tu s and the p r o c e s s e s of u rb a n iz a tio n and in d u s
t r ia lis m . He s ta te s th a t his patien ts w ere in m a n y w ays m a rg in a l
people at the onset of tu b e rc u lo s is and that d isin te g ra tio n of the p a tie n t’s
p re c a rio u s p sy c h o so c ia l a d ju stm e n t a lm o s t in v aria b ly o c c u r re d in the
tw o -y e a r perio d p reced in g o n set o r re la p s e of the d is e a s e . T hey w e re
“ se n sitiv e , anxious, rig id , and em otionally labile
H a rtz (30) d e s c rib e s th re e patients who d e m o n s tra te that an
individual m a y re a c t to life situ atio n s with an anxiety sta te o r o th e r
p e rso n a l b e h a v io r in su ch a way a s to in te rf e r e with healthy living; and
th ese re a c tio n s m ay b e c o m e a m o s t significant fa c to r in the on set and
c au se of tu b e rc u lo s is .
11
Psychological F actors and Irre g u la r Discharge
A num ber of recent studies investigate the relationship of p e r
sonality tra its to irre g u la r discharge. Moran, et al. (46), com pared
irre g u la r discharge patients with regularly discharged patients on the
basis of prehospital life adjustm ent and behavior within the hospital.
They concluded that hospitalization is but the m ost recent of a long
se rie s of life situations in which the irre g u la r discharge patient has
dem onstrated his inability to make an adequate adjustment.
Several studies using the Madison Sentence Completion Test (7)
were ca rrie d out. In one of these Newman and S p a re r (48) concluded
that inability o r unwillingness on the part of tuberculosis patients to
verbalize reactions to illness m akes them likely candidates for se lf
discharge. Conversely, m ere ability to verbalize reactions does not
enable people to adjust to hospitalization. The technique involved
m erely completing sentences and did not a sse ss the content of the
se n te n c e s.
Inspired by the Newman hypothesis, Kendig (37) also used p ro
ductiveness on the Madison Sentence Completion T est as the m easure
of verbal facility. Thirty-eight patients discharged with m axim al
hospital benefit (MHB), and a s im ila r num ber who left against medical
advice or were absent without leave (AWOL), were rated for their
gross expressiveness in term s of num ber of words completed and
num ber of words used. They w ere also rated for differential
12
expressiveness to sentence stem s of varying stim ulus value. She found
that the MHB group differed significantly from the AWOL group in that
it completed m o re sentences. The num ber of words p e r sentence
completed did not differ significantly for the two groups. T h e re was
negligible difference between the groups as far as ability to respond to
m o re emotionally stim ulating sentence ste m s. Finally, the frequency
distribution of total num ber of sentences completed shows the fam iliar
J curves of conformity, with the irre g u la r group showing less con-
form ity. Follow-up of the last study resulted in confirm ation that
nonconformity ra th e r than inability to verbalize is an im portant factor
underlying irre g u la r discharge. On the other hand, it appears that
inability to verbalize was confused in this study with unwillingness to
verbalize. No effort appears to have been m ade by Kendig to d iscover
how the two groups actually varied in true ability to verbalize.
Calden (7) also used the Madison te st to successfully d is c r im
inate between the two groups. Lewis, et al, (39), used the sam e test
and found that neuropsychiatric diagnosis, age, race, and distance
from hospital did not differentiate the two groups. S alazar and
Thurstone (52) observed that patients who leave p rem aturely a re e m o
tionally im m ature and cannot “ blow off s te a m .” V ern ier (59), using
the Rorschach and H o u s e -T re e -P e rs o n (HTP) T e sts, was able to p r e
dict type of discharge with an average accuracy of prediction of better
than 85 p er cent. H er groups w ere equated for such variables as age,
13
race, m a rita l status, socio-economic level, degree of disease, and
type of treatm ent.
Improvement Rate
The studies attempting to relate psychological factors to im
provement rate are relatively few and, as mentioned previously, are
for the m ost part poorly controlled o r inconclusive.
Basing their findings on psychoanalytically oriented psycho
therapy interviews with 16 patients, Benjamin, £t al. (3), concluded
that they were unable to find any personality features common to all
patients either as to general structure o r specific conflict prior to the
onset of the disease. However, they had the “ definite im pression . . ,
that inhibited hostility exerts an unfavorable influence while capacity to
express conscious hostility has a favorable influence. However, in
deeply re p re sse d hostility, the course was excellent.”
Wittkower, Durost, and Laing (64) found suggestive evidence
that emotional adjustment— other factors being equal— make for a
favorable course, while emotional m aladjustm ent influences it unfavor
ably. Patients were matched for age, sex, and cultural background.
However, the age range was 40 to 60 and the relative im portance of
emotional factors in this age group need not have been the same as for
a younger group. Forty patients having a favorable course were com
pared with the sam e num ber having an unfavorable course. The basic
14
techniques used were examination of the clinical history and interview
of the patient. Again, no uniform personality type was found, but
emotional conflict between obtaining satisfaction of the need to be loved
and cared for and the expression of aggression was characteristic.
The m anner of solving this conflict between dependence and aggression
varied from a frankly parasitic, demanding attitude toward the world to
a rigid rejection of dependency needs and an exaggerated striving for
independence. One sub-group was characterized as co n flic t-h arassed .
They are characteristically rigid, inhibited persons, h arassed by
feelings of guilt over sexual drives and hostile urges, the real source
of which they are unable to place and define. When faced by situations
arousing anger o r sexual excitation, appropriate action is inhibited,
creating states of chronic emotional tension. Of eight patients of this
kind, six had an unfavorable course despite only m oderately advanced
d is e a s e .
Rebner (51) used the Blacky Picture and Rorschach T ests to
investigate the role of personality as a factor in relapse in tuberculosis.
He hypothesized that oral and anal aggressive features would be m ore
prominent in relapsed groups than in unrelapsed ones. Also, he
hypothesized that oral and anal aggressive determ inants would be
directed toward the self in g re a te r degree in the relapsed groups. A
third hypothesis was that ego strength of the unrelapsed group would be
g re a te r. All except the last hypothesis were supported by the data at a
15
significant level. Now, it m ight s e e m th a t the f ir s t hypothesis c o n tr a
dicts the findings of B en jam in {see page 3) and C o lem an {see page 13).
H ow ever, the la tte r in v e stig a to rs had re fe re n c e to ho stility e x p re s s e d
in psychoanalytic in te rv ie w s, w h e re a s R eb n e r w as dealing with a g g r e s
sion as in fe rre d fro m te s t r e s p o n s e s .
U sing the R o rsc h a c h T e s t alone, Cohen (15) sought to p re d ic t
two y e a rs beforehand the m e d ic a l p r o g re s s ra te of young, m a le , h o s
pitalized p u lm o n ary tu b e rc u lo sis patients tre a te d e a rly in th e ir illn e ss .
None of the R o rsc h a c h v a ria b le s was found re lia b le to p re d ic t the ra te
of m e d ic a l p r o g r e s s . N or was d e g re e of m e d ic a l p r o g re s s significantly
re la te d to age, length of h o sp italizatio n , in tellig en ce, o r s e v e rity of
the tu b e rc u lo sis p r o c e s s .
An a tte m p t to d e riv e e m p iric a lly and v alid ate a sc a le fo r p r e
dicting rate of re c o v e ry in tu b e rc u lo sis was m a d e by B ro tm a n (6) who
used the M innesota M ultiphasic P e r s o n a lity In ventory and a few p h y sio
logical v a r ia b le s . The re s u lts w e re e s se n tia lly n e g a tiv e . It is
im p o rta n t to note that he did not c o n tro l fo r continuity of h o sp italizatio n
n o r for n u m b e r of d is c h a rg e s . F u r th e r m o r e , no co n tro ls for s i m i l a r
ity of m e d ic a l tre a tm e n t w ere in stitu ted .
M o ran (45) exam ined the re la tio n sh ip b etw een the ra te of
re c o v e ry of tu b e rc u lo sis patients and four psychological v a ria b le s:
(a) b eh av io r on the w ard, (b) attitudes tow ard the h o sp ital e n v iro n m e n t,
(c) p a st b eh av io r before h o sp italizatio n , and (d) fantasy a s m e a s u r e d by
16
the Them atic Apperception T e st. The rate of reco v ery from tu b e rc u
losis was estim ated for 46 patients by the amount of time required for
bacteriologic conversion to occur. V ariability in rate of reco v ery in
tuberculosis was found to be strongly associated with adaptive behavior
on the ward, m oderately associated with m e a s u re s of fantasy and
unassociated to ex p ressed attitude o r to prehospital behavior.
Pauleen (49) using a Q - s o r t technique found that patients who
made satisfacto ry m edical p ro g re ss tended not to o v e rre a c t emotionally
to the hospital situation. She also hypothesized a relationship between
personality factors and irre g u la r discharge which rem ained uncon
firm ed. She raised the question of the value of treating the h e te ro
geneous irre g u la r discharge group as a unity.
The R orschach Test, the B ender G estalt T est, and the House-
T r e e - P e r s o n T est w ere employed by V ern ier (60) in investigating
possible psychological factors in im provem ent rate. Even on c ro ss
validation fast im proving groups could be differentiated on the basis of
the following factors: little d e sire to be an invalid and to be passively
dependent, good relationships with other people and an absence of
anxiety o r tension. F u rth e rm o re , only m inim al changes w ere noted
when adm ission and discharge tests of both groups were com pared.
Sum m ary of L ite ra tu re
The m ain theme of the studies presented above is that
psychological facto rs a re to a g r e a te r o r l e s s e r d eg re e co n trib u to ry to
the onset of tu b e rc u lo sis. M ost w rite rs felt that em otional d isturbance
p rio r to hospitalization differed in kind and d eg ree. H ow ever, c o n
flicts involving dependency and a g g re ssio n w ere frequently m entioned.
Studies having to do with prediction of d isc h a rg e em p h asized c o n fo rm
ity and verb alizatio n ability as im p o rtan t fa c to rs . R e s e a rc h dealing
with im p ro v e m e n t ra te was by and la rg e inconclusive.
Rationale of P r e s e n t Study
The review of the lite ra tu re p re se n te d suggests that psycholog
ical fac to rs a re im p o rtan t in the onset and co u rse of tu b e rc u lo sis.
F u r th e r m o r e , the lite ra tu re su g g ests that the relationship of p sy ch o
logical fac to rs to im p ro v e m e n t ra te can be c le a rly identified only if
o th e r nonpsychological facto rs such as age, kind of tre a tm e n t, and
d eg ree of pathology a re controlled. The p re se n t study is designed to
te s t the hypothesis that a c c e ssib ility to r e p r e s s e d a r e a s of em otional
d istu rb an ce is related to im p ro v em en t r a te . As a lre a d y m entioned, the
e x p e rim e n ta l factor is m e a s u re d in te r m s of ability to v e rb a liz e to
r e p r e s s e d conflict a r e a s and also in te r m s of the n u m b er of such
a r e a s .
The concept of a ccessib ility to r e p r e s s e d em otional d isturbance
and its relatio n to so m atic change through the expenditure of en erg y
a r e d erived from the psychoanalytic fra m e of re fe re n c e . Hence it is
18
ap p ro p ria te to s u m m a riz e b riefly psychoanalytic thinking with r e g a rd
to the concepts involved in the definition of a c cessib ility : r e p re s s io n ,
neurotic conflict, elaboration of d eriv a tiv e s, and expenditure of
en erg y .
A ccording to F en ich el (27), a neurotic conflict is defined as one
betw een a tendency striv in g for d isc h a rg e and another tendency that
trie s to prev en t this d isc h a rg e . Such a conflict is fu rth e r c h a r a c t e r
ized by lack of a w a re n e ss o r r e p re s s io n . P u t another way, an im pulse
o r drive is unacceptable b ecau se of the anxiety o r o th er painful effect
which would be a ro u se d if the drive w ere e x p re ss e d . H ow ever, the
im pulse is n e v e r totally r e p r e s s e d and is frequently e x p re ss e d in a
disguised fo rm (the deriv ativ e). Again, the disguise o r d erivative
often does not suffice for adequate e x p ressio n of the im p u lse. A c c o rd
ing to D ittm an and Raush:
As im pulse stren g th in c r e a s e s , the g r e a te r will be the tendency
tow ard d ire c t im pulse e x p re ssio n and the g r e a te r the en erg y
re q u ire d to inhibit d ire c t e x p re ssio n . In o r d e r to m aintain
equilibrium in the sy ste m , energy m u st be expended to prevent
panic. The en erg y bound up in this p ro c e ss , w-hich has been
called c o u n te r-c a th e x is, is not available for o th er o rg a n ism ic
functions. (23:3 88)
In addition, it is to be noted that when the im pulse is blocked,
id eas, affects, and perceptions of outside s tim u li asso c ia te d with it
also m ay undergo r e p re s s io n and disto rtio n . The p ro g re ssiv e dev elo p
m en t of this p ro cess has been r e f e r r e d to as the layering of defenses.
As this defensive s tru c tu re becom es m o re com plicated, m o re energy
19
is r e q u ire d to m ain tain the s tr u c tu r e itse lf and m o re a r e a s of living
a r e p e rm e a te d by psychopathology.
T he way in which the above d isc u ssio n is re la te d to the concept
of a c c e s sib ility b e c o m e s m o re m eaningful in the context of p sy c h o
th e ra p y . Kubie (38) in d icates that the p r o c e s s of p sy ch o th e rap y is
co n ce rn e d with m aking so m e of the r e p r e s s e d conflict a r e a s conscious
in o r d e r th at the patient le a rn to reco g n ize, to le ra te , and p e rh a p s
elim in a te the anxiety connected with the im p u lse . In th is way the
patient w ill be m o re c o m fo rtab le and will have available to him en erg y
“ bound u p ” in the conflict a r e a . G e n e ra lly speaking, a c c e s s ib ility
has r e fe r e n c e to the e a s e with which a patient can beco m e a w a re of
r e p r e s s e d conflict a r e a s . It a p p e a rs logical th at if m u ch e n e rg y is
bound up in the defensive p ro c e s s as when the a r e a s of conflict have
been g re a tly e la b o ra te d , the p sy ch o th e rap eu tic ta sk will be m o re diffi
cult and the patient is said to have le ss a c c e s sib ility . D o llard and
M il l e r ’s sta te m e n t with r e g a rd to th e ra p y p ro g n o sis is a p p ro p ria te
h e re :
T he s tr o n g e r the ego, the b e tte r the p ro sp e c ts for r e c o v e ry .
. . . The p erso n with s tro n g adaptive b eh av io r in c e rta in a r e a s of
his life shows that he has th ese g e n e ra l e le m e n ts to build on. F o r
this re a s o n even s e v e r e conflict lim ited to one a r e a is m o re
fav o rab le than g e n e ra l r e ta rd a tio n of the whole d evelopm ent.
<24:236)
In the above d isc u s s io n , groundw ork w as laid fo r a definition of
one of the m e a s u r e s of a c c e s sib ility used in this study. T h is definition
20
m ay be ex pressed thus: accessibility v aries inversely with the num ber
of re p re sse d conflict a re a s .
The definition of a second m easu re of accessibility is derived
from the psychoanalytic technique of free association. According to
Dittman and Raush (23), the method of free association is used in
psychoanalysis because free associations are highly determ ined by
unconscious m otivations. Also these associations are supposedly
rep resen tativ es of unconscious em otional conflicts and express the
unacceptable im pulses in a disguised form .
Yet re sista n c e to the flow of free association n e c e ssa rily occurs
when the anxiety connected with the re p re sse d conflictual m a te ria l is
close to aw aren ess. Patients m ay block or adopt other evasive and
defensive m e a s u re s . Thus, they v ary in the ability to free associate
in the face of anxiety presum ably according to the strength of the
im pulse, and the energy available for defending against it. In psycho
analytic therapy it is the ability to free associate which is the
im portant index of the outcome of the case. In this connection, Kubie
(3 8) re m a rk s that the freedom with which such associations keep flow
ing is one of the im portant indicators of p ro g re ss in psychotherapy.
On this basis, then, accessibility to re p re sse d emotional disturbance
m ay be m easu red in te rm s of ability to free associate to such a re a s.
However, a possible contradiction to the above posited relatio n
ship could be discussed. It might be argued that ability to free
21
associate to re p re sse d conflict a reas is not a good prognostic indicator
in the case of people who tend to intellectualize and use verbalization
as a ch aracteristic defense m echanism (£. g. obsessive-com pulsives,
compulsive personalities, etc.). According to psychoanalytic thinking,
intellectualization as used by such patients involves isolation of effect.
Yet it is characteristic of the m ore seriously ill among these groups
that affect is not altogether isolated and anxiety can em erge. If
anxiety does threaten to break through, intellectualization and v e rb a l
ization are restricted; blocking or rep ressio n and consequent slowing
of the flow of free association often occurs. The patient m ay block
because of the need to find the precise word o r because he m ay require
m o re o rd e r in his thinking than is afforded by the technique of free
association. Even in the case of the m ore “ n o rm a l” intellectualizer,
there may be little true productivity of associations: constant r e p e
tition o r verbalization about everything but the conflict a re a s may take
place. It is questionable, then, whether it is c o rre c t to assum e that
verbalization on the part of these groups contradicts the prem ise that
accessibility to re p re sse d areas of emotional disturbance is directly
related to im provem ent in psychotherapy.
In the preceding discussion the psychoanalytic approach was
employed to derive the concept of accessibility and to relate it to
im provem ent in psychotherapy. The sam e approach also implies a
relationship between accessibility and somatic changes when it re fe rs
22
to energy being “ bound up in . . . counter-cathexis” and consequently
unavailable for other organism ic functions. Among others, Dunbar
(26), W eiss and English (61), and Alexander (1) have w ritten extensive
ly on the relationship of psychological factors to physical illness.
While they may differ as to how psychological factors contribute to
somatic changes, they do agree that psychotherapeutic treatm ent will
tend to alleviate physical sym ptom s to the extent that they a re r e v e r s
ible and connected with emotional factors. In tuberculosis many
w orkers believe that the m ore energy bound up in psychological
conflict and control of anxiety the less there is available for the healing
process. Wittkower (63) feels that re p re sse d emotional conflict
lowers the resistance of the body to the invasion of the tubercle bacilli.
Selye (53) and Clarke (12) have shown that any s tre s s (including acute
and chronic anxiety) can result in profound pathological physiological
changes via the adrenal cortex. M oreover, Holmes, et al. (32), have
shown that the effect of life s tre s s (emotional conflict) plays a role in
the course of tuberculosis .
The psychoanalytic approach may then be applied to the psycho
som atic a re a in this way: to the degree that somatic pathological
p ro cesses are concomitant with o r exacerbated by re p re sse d emotional
conflicts, such somatic conditions in so far as they are reversible
should be relieved or eliminated as therapy p ro g resses. Following the
reasoning applied above, the m o re accessibility to begin with, the m ore
23
rapid the physical im p ro v e m e n t.
S tatem ent of the P ro b le m
While th e re is g e n e ra l feeling that tu b e rc u lo sis is m u ltic a u sa l,
a u th o ritie s also feel that psychological fa c to rs a r e im p o rta n t in the
o n set and c o u rse of tu b e rc u lo sis. It is im plied that b e tte r o v e r -a ll
em otional a d ju stm en t is re la te d to im p ro v em en t. H ow ever, v e ry few
psychological o r m e d ic a l studies of a sy ste m a tic , controlled nature
have been designed o r proved to be significant.
The p re se n t study is an attem p t to investigate the relationship
betw een a c c e ssib ility to r e p r e s s e d em otional conflicts and im p ro v e m e n t
ra te in tu b e rc u lo sis.
The assu m p tio n s on which this study is b ased a re :
1. The prognosis fo r psychotherapy is b e tte r the m o re the
patient can fre e a sso c ia te to r e p r e s s e d a r e a s of em otional d istu rb a n c e .
2. The prognosis is also b e tte r the few er the n u m b er of
r e p r e s s e d conflict a r e a s .
3. P h y sic a l sym ptom atology in so fa r a s it is concom itant
with, ex acerb ated by, o r contributed to, by em otional d istu rb a n c e , w ill
tend to be alleviated m o re quickly am ong patients who re v e a l g r e a te r
ability to fre e a sso c ia te to a r e a s of r e p re s s e d d istu rb an ce o r who have
few er r e p r e s s e d conflict a r e a s .
24
H ypotheses
In g en eral, in pulm onary tu b e rc u lo sis, if n on-em otional fac to rs
a re controlled, ac c e ssib ility to r e p r e s s e d em otional d istu rb an ce w ill
v a ry d ire c tly with im p ro v em en t ra te .
M ore specifically it is hypothesized that:
1. A ccessib ility m e a s u re d by ability to fre e a sso c ia te to
r e p r e s s e d a r e a s of em otional d istu rb a n c e will tend to be positively
re la te d to im p ro v em en t ra te .
2. A ccessib ility m e a s u re d by l e s s e r n u m b e r of r e p re s s e d
a r e a s of em otional d istu rb an ce will tend to be positively related to
im p ro v e m e n t r a te .
The te s t of the hypotheses will be acco m p lish ed by m aking the
e x p e rim e n ta l groups as hom ogeneous as possible on the following
fa c to rs: type of m edication, continuity of hospitalization, absence of
p rio r hospitalization, age, sex, extent of d ise a se , ab sen ce of oth er
s e rio u s d ise a se , and no evidence of s e rio u s em otional d istu rb an ce.
C H A PT E R III
Method
In this study it is hypothesized that among tuberculosis patients
accessibility to re p re s se d emotional disturbance is one of the factors
related to recovery tim e .
In this chapter will be described (a) the patient population,
(b) the instru m en t used, {c) the procedures for gathering the data
involved in testing the hypotheses, and (d) the m e a su re s devised to
te st quantitatively the hypotheses.
Subjects
Subjects w ere m ale veterans at San Fernando and Long Beach
V eterans A dm inistration H ospitals, with a diagnosis of pulmonary
tu berculosis. They w ere divided into two groups on the basis of extent
of pathological involvement of the lung at the tim e of adm ission to the
2
hospital. Consultation with m edical personnel resulted in defining the
le s s e r pathology group as one including patients who bore a diagnostic
classification of pulmonary tuberculosis, m inim al, or m oderate without
cavitation. The g re a te r pathology group included those who were
classified as pulmonary tuberculosis, far advanced, o r m oderate with
cavitation.
2
D rs . C. A nderson and C. Smith, Long Beach VAH.
26
T h e re w ere 31 subjects with l e s s e r and 33 with g r e a te r path o l
ogy. T h e ir ages ranged fro m 20 to 40 y e a rs with m edian ages of 25
and 29 y e a r s for the l e s s e r and g r e a te r pathology groups, resp ec tiv e ly .
Education for the g r e a te r pathology group v arie d from nine to sixteen
y e a rs with 82 p e r cent of the subjects having m o re than twelve y e a r s .
F o r the l e s s e r pathology group y e a rs of education varied fro m nine to
16 y e a rs with 90 p e r cent of the su b jec ts having m o re than 12 y e a r s .
3
Intelligence of all subjects was av erag e o r above. Seventy p e r cent
of the l e s s e r pathology group a re C aucasian; 80 p e r cent of the g r e a te r
pathology group a r e C au c asian (Table 1),
A ll su b jects sele cted w ere s im ila r in that they had been
continuously hospitalized, had had no o th e r m a jo r d ise a se while in the
hospital, and had not undergone s u r g e r y as p a rt of tre a tm e n t. All had
4
had equivalent ch em o th erap y . None had been hospitalized for this
d ise a se b e fo re . A ll patients were se m i-a m b u la to ry , C la ss IV patients.
This is to say d ise a se activity had been definitely halted as shown by
the fact th at all revealed negative findings in bacterio lo g ic and o th er
3
Many of the subjects had not had fo rm a l intelligence te s ts .
The s ta te m e n t above is based on psychological s c re e n in g interv iew s,
on those fo rm a l te sts which w ere a d m in iste re d by psychologists, and
on the e s tim a te of intelligence m ade by the staff psychologist at the
o v e r -a ll evaluative staff conference held for each patient at the
hospital at various stag e s in the c o u rse of tre a tm e n t.
4
M edical perso n n el ag reed that any two of the th re e drugs
(strep to m y c in , PAS, o r INH) a re equivalent in healing value.
27
Table 1
Education, Age, Occupational Status,
and Ethnic Group of the
G r e a te r and L e s s e r Pathology G roups
V ariable
G r e a te r Pathology
Group
(N = 33)
L e s s e r Pathology
Group
(N = 31)
Median Range N Median Range N
Education
(years)
12.5 9-16 12.5 9-16
Age
(years)
29.0 20-40 25.0 20-40
Occupational
Status
M anual 17 13
White C ollar 16 18
Ethnic Group
C aucasian
Negro
Mexican
27
2
4
22
4
5
28
la b o ra to ry te s ts for a period of at le a st five m onths (22).
O ccupational statu s w as difficult to a s c e rta in read ily in c a se s
of the younger subjects who had gone d ire c tly fro m school into the
s e rv ic e w here they had in c u rre d the d is e a s e . H o w e v e r, it was possible
to classify even these subjects into “ m a n u a l'’ and “ white c o lla r ” on
the b a s is of the p ro g ra m followed in school o r during the a rm e d s e r
vice , o r p reviously planned future tra in in g p ro g ra m o r occupation.
The relatio n sh ip of occupational status to the e x p e rim e n ta l v a ria b le s in
5
this study w as investigated and found to be nonsignificant.
No psychotic o r o th erw ise se rio u sly em otionally d istu rb ed su b
je c ts w ere included in the study becau se they m ight not have been
c o - o p e ra tiv e .
E v e ry patient who could m e e t the above e x p e rim e n ta l c r ite r ia
w as included in the study. H ow ever, since such subjects form a v ery
sm a ll proportion of the total C la s s IV population at any given tim e, a
period of a p p ro x im a te ly one y e a r was re q u ire d to s e c u re the n u m b er of
patients who m ake up the p re se n t sa m p le s of both g roups. Some q u e s
tion m ight be ra ise d with r e g a rd to having tested the subjects at the
tim e they reach ed s e m i-a m b u la to ry sta tu s. It m ight be argued that
a c c e ssib ility as m e a s u re d m ight not have been the sa m e if the subjects
had been tested e a rly in th e ir hospitalization. H ow ever, Muldoon (47),
5
See page 51
29
B e rg e r (4), and C harens (11) have dem onstrated that following an in i
tial s tr e s s reaction, patients quickly re v e rt to th e ir prehospitalization
personality.
Instrum ents
The Jung Word Association T est (35) was selected as the basis
of the m e a s u re s of accessibility to emotionally disturbing a r e a s . The
Jung T est was chosen ra th e r than any oth er word association test
because relatively few of the w ords composing it seem to be capable of
arousing em otional response on the part of m ost people; they have no
gro ssly sexual o r vulgar m eaning. On the whole, they a re frequently
used in m agazines, popular literatu re, and general conversation. It
would seem , then, that if any of these words did stim ulate a disturbing
response, it might be m o re validly assu m ed that it had really tapped an
a re a of special significance for the subject.
The indicators of emotional disturbance on this test as outlined
by Jung a re well known. In this study the following complex indicators
a re used:
1. Reaction Tim e .
2. Repetition of stim ulus w o rd .
3. F ailu re to respond to stim ulus word o r claim of inability to
h e a r it.
4- Changed response to stim ulus word on second t r i a l .
30
A stopwatch capable of recording reaction tim e to the n e a re st
tenth of a second was used.
The w rite r used as his criterion of emotional loading a com bi
nation of long reaction tim e and one of the other three complex
in d ic a to rs .
It is generally held (67,62) that reaction tim e is one of the m ore
reliable and valid indicators of disturbance. Woodworth and Schlos-
b erg (67), in reviewing the literatu re on word association te sts, imply
that a reaction tim e of two seconds or m o re may be considered indica
tive of emotional disturbance. In this study 20 per cent o r less of the
reaction tim es on each protocol equal o r exceed two seconds.
C ritic ism s of word recognition studies which m ay also be
applicable to reaction time as used here w ere first m ade by Howes and
Solomon (33) and P ostm an and Schneider (50). They held that diffi
cult m eanings o r infrequent usage might be the re a so n s for inability to
recognize taboo w ords and that rep re ssio n o r perceptual defense was
not the b asis for such inability. However, while DeLucia and Stagner
(19) found a significant relation between word frequency and visual
recognition time, they found no significant trend between frequency and
word reaction tim e.
At any rate, in o rd e r to m inim ize the im portance of word
difficulty in this study, each subject, following the testing, was asked
31
the m ean in g of each w ord on the pro to co l with a re a c tio n tim e o v e r two
seco n d s. W ords which w ere not e a sily defined w ere o m itted fro m
s ta tis tic a l tre a tm e n t. T h e n u m b e r of such w o rd s, h o w ev er, proved
m in im a l. F re q u e n c y of usage w as found to be u n im p o rtan t in this
study for when the T h o rn d ik e -L o rg e (57) freq u en cy w ord lists w ere
u sed , it w as found that the m a jo rity of the Ju n g stim u lu s w ords w e re
am ong the 1,000 m o st frequent w o rd s and none of the r e m a in d e r w e re
infrequently o c c u rrin g w o rd s.^
Two com plex in d ic a to rs a r e used on the b a s is of Z i m m e r ’s (68)
study on the relatio n sh ip of se lf-a c c e p ta n c e to conflict. He u sed tr a it
ad jectiv es a s stim u lu s w o rd s on a w ord a sso c ia tio n te s t. Six of the
tra d itio n a l com plex in d ic a to rs w e re obtained and em ployed a s indexes
of conflict. He found that tr a its with no in d ic a to rs , a s w ell a s those
which evoked two o r m o re com plex in d ic a to rs , d ifferen tiated s u c c e s s
fully betw een two gro u p s of d istin ctly d iffe re n t em otional a d ju stm e n t,
while tr a it s with one com plex in d ic a to r did not. Hull and Lugoff (34)
a lso indicated that coincidence of two o r m o re com plex in d ic a to rs on
one stim u lu s w ord co n sid e ra b ly i n c r e a s e s the likelihood that the
stim u lu s w ord has touched upon an em otionally im p o rta n t com plex.
A ffective Rating Scale
It h a s a lre a d y been stated that a c c e s s ib ility h as r e fe r e n c e to
See Appendix.
32
re p re s s e d a r e a s of em otional d istu rb an ce, i_. e., the subject is unaware
of these a r e a s . In this study the o p erational definition of this variable
involves the selection of the Jung stim ulus w ords which a re emotionally
loaded but which at the sam e tim e a r e judged by the subject to be not
d istu rb in g .
Such judgm ents w ere obtained by requesting each subject to rate
the 99 stim ulus w ords as to d isturbance o r unpleasantness on a scale
of four steps with zero rep re se n tin g least intensity, and t h r e e ,
g re a te st intensity of d isturbance.
P ro c e d u re
Each subject w as seen for at le a st one h alf-hour before the
beginning of the ex p erim en t for the purpose of obtaining co-operation
with the e x p e rim e n te r. The patient was told that the p re se n t e x p e ri
m ent was undertaken so that the place of em otions and feelings in
tu b ercu lo sis might be b e tte r understood. Without exception, all
patients ag re e d that em otional facto rs a re im p o rtan t in the course of
the illn e ss. In fact, a n u m b er of the patients then took the opportunity
of citing instances during th e ir hospital stay when em otional conflict
seem ed to be in terrupting o r slowing th eir own p ro g re s s o r that of
other patients. Many took the opportunity of using the period for the
purpose of ex p re ssin g th e ir fe a rs and w o rrie s about the future as in a
kind of psychotherapy sessio n .
33
Following this motivation period each subject was given the
following instructions:
P le a se relax and stretch out in your chair. Now close your
eyes until I tell you to open them . I’m now going to say a list of
words. As soon as I say one word, you say the first word that
com es to your mind a s quickly as you can. It’s im portant that you
say your word a s quickly as possible. Any word you say will be
all right. Your word doesn’t have to have anything to do with the
one I say. I won’t be shocked by anything you say, either. The
im portant thing is to say the first word that com es to your mind
as fast as you can.
The 99 stim ulus words w ere presented and the reaction time
was obtained to the n e a re st tenth of a second by the stopwatch. The
tim e was noted and the responses were recorded in shorthand.
The stim ulus words w ere given a second tim e and the subject
was instructed to try to give the sam e response to each word. After
the second presentation, the following was said:
It m ay be that some of the words I told you w ere unpleasant,
disagreeable, o r disturbing. It m akes no difference why you felt
the way you did about the w ords. It could be they rem inded you of
something; it may be the sound itself was disturbing o r unpleasant;
or it m ay be you have no idea why you disliked the word. At any
rate, when I say these words now, tell m e how you feel about these
words in this way: if the word is not disturbing o r unpleasant at
all, say z ero ; if a little disturbing, say one; if m oderately d is
turbing, say two; if the word is v ery unpleasant say t h r e e .
Selection of N eutral and Em otionally Loaded W ords
The subject was next asked to “ stretch a b it” while the e x p e ri
m en ter search ed his protocol for the words which reflect “ emotional
disturbance of which he is u n aw are.”
34
Selection of Em otionally Loaded W ords
F irs t, the words which w ere rated “ z e ro ” by the subject but
which had a reaction tim e of two seconds o r m o re w ere sought out.
7
Of these, five words were selected which in addition to reaction
time bore one of the following “ complex in d icato rs” :
1. Repetition of the stim ulus w ords.
2. Changed response on the second trial.
3. F ailu re to respond to stim ulus word or claim of inability to
h ear it.
Selection of N eutral W ords
A nu m b er of words equal to that of the emotionally loaded words
was selected as the control. T hese w ere words rated as not disturbing
(zero) by the subject, with reaction tim es of one second o r less, and
g
which c a rrie d none of the complex indicators.
F re e A ssociation to Emotionally Loaded and N eutral W ords
The subject was then instructed as follows:
If, as frequently happened, there w ere m o re than five words
available for selection, those chosen were the first having the longest
reaction tim e. If reaction tim es were equally long, then g reatest
num ber of other complex indicators was the basis of selection.
Among approxim ately 30 p e r cent of all subjects, only four
words could be found. H ere, the four words were treated as sc o re s
in the sam e way as described for five.
g
W henever possible, each of the neutral words was selected
from successive groups of 20 on a protocol. In all c a se s, the first
word in each group m eeting the c rite ria was the one chosen.
35
P le a se keep your eyes closed as before. I a m going to say
som e w ords again. Now, how ever, when I say a word, say as
m any single w ords as you can until I tell you to stop. It m akes no
difference what w ords you say, but say them as quickly a s you
can and try to say single w o r d s .
The ten w ords w ere then presented; emotionally loaded words
alternated with n eu tra l w ords. The subject was allowed to free
asso ciate for one m inute to each w ord. The words w ere reco rd ed in
shorthand.
M e a su re s and Anticipated Results
A ccessibility to r e p re s s e d em otional disturbance is m e a s u re d in
two w ay s:
1. Each patient’s accessib ility is e x p re sse d in te r m s of a
ratio in which the n u m e ra to r is the sum of all the w ords he associated
to his five em otionally loaded w ords; the denom inator is the grand sum
of the w ords asso ciated to five n eu tral w ords and the five loaded w ords.
T his ratio (L/N+L.) is used as a sco re for each subject.
The need for such a ratio becom es evident when it is considered
that differences in v erb a l ability a re attributed to m any facto rs other
than em otional disturbance (e .g ., linguistic background, education,
reading, idiosyncratic physical factors). Hence, a baseline for each
subject m u st be set up in o r d e r that he be com pared with him self, so
to speak. The baseline is the denom inator of the above ratio.
2. The second m e a s u re of accessib ility is derived fro m the
the sam e instrum ent. The total num ber of loaded words rated zero is
recorded for each subject and also constitutes an accessibility score.
Rate of P h y sical Im provem ent
This variable is m easu red in te r m s of the tim e elapsed from
the initiation of continuous hospitalization and uninterrupted ch em o
therapy until the subject has reached sem i-am b u lato ry status (Class
IV).
Applying these m e a s u re s , the following resu lts were anticipated
in te rm s of the m ain hypothesis: if non-em otional factors are con
trolled, then (a) the ratio s c o re s of accessibility (L/N+L) will vary
directly with im provem ent rate, and (b) the num ber of loaded words
rated zero will vary inversely with im provem ent rate.
C H A PTE R IV
Results
This chapter includes the statistical description of the
v ariab les, the re su lts of the statistical analyses of the data concerned
with the experim ental variab les, and certain supplem entary data.
Since the distributions of the two m e a s u re s of accessibility are
badly skewed and do not m eet the crite rio n of hom oscedasticity (28),
the appropriate test of the hypotheses is a n o n -p aram etric one. In all
except one case, the statistic selected was the m edian te st of Mood (43)
employing chi square. In one case where the expected cell frequencies
were less than five and the total num ber of subjects involved was less
than 20, F i s h e r ’s Exact P robability test was used. The one-tailed
alternative is appropriate since the hypothesis predicted the direction
of association of the v ariab les.
Frequency distributions for all variab les w ere split at the
median and a four-fold contingency table was set up in each case on
this b asis. F o r example, a frequency distribution of im provem ent is
p repared for the le s s e r and g re a te r pathology groups. The median was
determ ined and subjects whose time exceeded the median w ere d e sig
nated as fast improving and those beneath it as slow improving. In the
sam e way, a distribution of accessibility ratios is made and those
sc o re s above the median a re designated as high accessibility and those
38
below, as low accessibility. In all cases where chi square was used,
Y ates' co rrectio n for continuity was applied, for no theoretical cell
frequency was g re a te r than 25 (54).
Next, according to H olm es, et al. (32), occupational status is
related to onset and cause of tuberculosis. Since this variable could
not be controlled conveniently in this study, the relationship of o ccu
pational status to each of the experim ental v ariables was analyzed
using the m edian test and chi square as outlined above in the d e s c rip
tion of the statistical analyses of the experim ental v ariables.
Lastly, supplem entary data resulting from consideration of an
alternative to the m a jo r hypothesis of the present study is included.
Mahl (40) has indicated im provem ent will be slowed re g a rd le ss of
whether em otional conflicts a re unconscious o r not so long a s th eir
effect is to produce chronic anxiety. As applied to this study, his
hypothesis im plies that num ber of conflict a re a s alone (irresp ectiv e of
verb al rating of subjective disturbance) is related to im provem ent. It
is also conceivable that nu m b er of disturbing a re a s not re p re sse d , L
loaded w ords rated as disturbing, m ay be related to im provem ent. In
analyzing the data with reg ard to these hypothetical relationships, the
median test and chi square w ere employed as statistical tests in the
sam e m an n er as described above for the experim ental v ariab les.
However, the two-tailed alternative was used since the direction of
association was not predicted.
39
Statistical D escription
Consideration of the frequency distributions of im provem ent
tim e for the le s s e r and g re a te r pathology groups (Fig. 1) reveals that
72 per cent of the g re a te r pathology subjects fall within the range of
im provem ent tim e of 90 per cent of the le s s e r pathology subjects.
This overlap suggests that extent of pathology alone is not the only
factor involved in im provem ent tim e. However, when groups of s im i
la r d e g rees of pathology from the two hospitals w ere combined (see
Table 2), a significant difference w as found (£ = .01).
In o rd e r to determ ine w hether the corresponding groups were
appropriately combined, an analysis of the m ean im provem ent ra te s
of all hospital subgroups was c a rrie d out (Table 2). The g re a te r
pathology groups at the two hospitals did not differ significantly
(£ = .20). F u rth e rm o re , it m ay be seen in Tables 3 and 4 that the
m eans of both m e a s u re s of accessibility for these groups did not differ
significantly (for L./N+L, £ >.10; for loaded words rated zero, £ > .1 0 ).
Hence, the combination of the two g re a te r pathology groups appears
justified.
In the case of le s s e r pathology, it is seen (Table 2) that the
m ean im provem ent time of the SFVAH group was significantly g re a te r
than that of the LBVAH (£ = .001). F u rth e rm o re , these groups also
differed in that the m ean of one of the m e a s u re s of accessibility
(loaded words rated zero) was significantly higher for the SFVAH
Number of Subjects
40
G re a te r Pathology (N = 33)
L e s s e r Pathology (N = 31)
5-i
4-
3-
2 -
1-
300 600 700 100 200 400 500
Time in Days
Fig. 1. Frequency distribution of time elapsed from adm ission
date until reaching C lass IV for le s s e r and g re a te r pathology groups.
41
T ab le 2
Significance of D ifferences between M eans of Im provem ent Tim e
M e a su re d by T im e from Beginning of H ospitalization
to Date of Reaching S em i-am b u lato ry Status
(C la s s IV) for all Groups
Hospital
D egree
of
Pathology
N
Mean
(Days)
cr
d M
^ M
t
£
SFVAH
LBVAH
G r e a te r
15
18
376.0
307.1
106.0
159.2
69.1 48.0 1.4 <.20
SFVAH
LBVAH
L e s s e r
11
20
327.8
179.3
90.1
77.5
148.5 31.9 4.7 .001
SFVAH
LBVAH
SFVAH
LBVAH
G r e a te r
L e s s e r
33
31
338.3
231.8
142.3
105.2
106.5 31.8 3.3
>.01
LBVAH G r e a te r 18 307.1 159.2
20.8 53.0 0.4 .50
SFVAH L e s s e r 11 327.8 90.1
SFVAH G r e a te r 15 376.0 106.0
48.2 46.0 1.0 >.20
SFVAH L e s s e r 11 327.8 90.1
LBVAH G r e a te r 18 307.1 159.2
128.0 40.0
•5 9
<.01
LBVAH L e s s e r 20 179.3 77.5
O
42
T able 3
Significance of D ifferences between M eans of A ccessibility Scores
M easured a s the Ratio between N um ber of W ords A ssociated
to Loaded W ords Rated Z e ro , and the Sum of W ords
A ssociated to Both Loaded and N eutral W ords (L/N+L)
for All Groups
H ospital
D egree
of
Pathology
N
Mean
(L/N+L)
cr
DM
^ M
t
£
SFVAH
LBV AH
G re a te r
15
18
.400
.423
.088
.056
.023 .023 1.00 >.10
SFVAH
LBV AH
L e s s e r
11
20
.414
.422
.089
.054
.008 .002 .40 >.50
SFVAH
LBV AH
SFVAH
LBV AH
G r e a te r
L e s s e r
33
31
.416
.419
.072
.068
.003 .017
.18 >.50
LBV AH G re a te r 18 .423 .056
.009 .026 .35 >.50
SFVAH L e s s e r 11 .414 .089
SFVAH
SFVAH
G re a te r
L e s s e r
15
11
.400
.414
.085
.089
.014 .035 .40 >.50
LBV AH G r e a te r 18 .423 .056
.001 .021 .005 .90
LBV AH L e s s e r 20 .422 .054
43
T ab le 4
Significance of D ifferences between Means of A ccessibility Scores
M easured as the N um ber of W ords Loaded but Rated Zero
in Disturbance by Subjects for All Hospital Groups
Hospital
Degree
of
Pathology
N
Mean
Loaded
W ords
cr
d m
ffdM
t
P
SFVAH
LBVAH
G re a te r
15
18
9.70
7.55
4.62
3.53
2.15 1 .42 1.510 7.10
SFVAH
LBVAH
L e s s e r
11
20
9.91
5.80
5.40
3.58
4.11 1.61 2.560 .02
SFVAH
LBVAH
SFVAH
LBVAH
G re a te r
L e s s e r
33
31
8.53
7.26
4.15
5.15
1.27 1.66 .765 .50
LBVAH
SFVAH
G re a te r
L e ss e r
18
11
7.55
9.91
3.53
5.40
2.36 1.64 1.400 7.10
SFVAH G re a te r 15 9.70 4.62
0.21 1.90 .111 .50
SFVAH L e s s e r 11 9.91 5.40
LBVAH G re a te r 18 7.55 3.53
1.75 3.60 .486 .50
LBVAH L e s s e r 20 5.80 3.58
44
group, £ being .02 (Table 4). H ow ever, the m eans of the other
m e a s u re (L/N+L) did not show significant difference for £ is .50 (Table
3). T h e re fo re combination of the le s s e r pathology groups at the two
hospitals is open to som e question. T his is to say it is n e c e s s a ry to
analyze each of these groups sep arately when testing the hypotheses of
this study.
C onsideration of the SFVAH l e s s e r pathology group suggests
that the la tte r m ay be combined with the two g re a te r pathology groups,
for none of these th re e subgroups differed significantly in im provem ent
tim e o r on the two accessib ility m e a s u re s .
It is, then, only the LBVAH l e s s e r pathology group which is
significantly different from the o th ers in im provem ent ra te . A p o s si
ble explanation of this difference might be that the le s s e r pathology
patients at Long B each were fa r less physically ill than those at San
Fernando.
It is also in tere stin g to note that in alm o st all subgroups the
n u m e ric a l values of the m eans of both accessib ility m e a s u re s suggest
higher a c cessib ility for the groups of patients who have lower m ean
im pro v em en t tim e s. It is to be em phasized, how ever, that only one
of these differences is significant.
The frequency distributions of the L/N+L s c o re s (Fig. 2) for
both degree of illn ess groups a re negatively skewed. As was the case
with the im pro v em en t time c u rv es, it is seen that there is considerable
Number of Subjects
45
G re a te r Pathology Group (N = 33)
L e s s e r Pathology Group (N = 31)
5-i
4 -
1 -
.10 .20 .30 .40 .50
L/N+L Ratio S cores
Fig. 2. F requency distributions of accessib ility ratio sc o re s
(L/N+L) of L e s s e r and G r e a te r pathology groups.
46
overlap between the accessibility sco re s of both groups. The highest
accessibility score in each group does not exceed .50, indicating that
free association to loaded words rated not disturbing did not exceed
that to neu tral w ords. The frequency distribution of num ber of loaded
words rated zero (Fig. 3) is skewed positively in the case of the le s s e r
pathology group, while that of the g re a te r pathology group appears
m o re sy m m etrical. H ere again the overlap of sc o re s is very great.
As pointed out above, the skewness of the distributions req u ires that
n o n -p aram etric techniques be employed.
Hypotheses
A ccessibility Ratio (L/N+L) in Relation to Im provem ent Tim e
The hypothesis stated that provided non-em otional factors are
controlled, the ratio sco res will vary directly with im provem ent r a t e .
In Table 5 it is seen that in the g re a te r pathology group the
statistical findings support the above hypothesis that ratio sco re s are
related to im provem ent ra te . Chi square is 5.18, which is significant
at the .025 level. However, the le s s e r pathology group reveals no
trend, p being between .80 and .90.
When the two g re a te r pathology groups a re combined with the
San Fernando le s s e r pathology group, chi square is 4.79 and £ is b e
tween .02 and .05. This is also in the direction predicted by the
hypothesis.
47
G re a te r Pathology (N = 33)
L e s s e r Pathology (N = 31)
6 - i
5-
w
C J
< u
* > — 3
X
3
CO
3-
o
u
0 )
X)
a
3
£
1-
X
4 12 20 0 8 16
A ccessibility Scores (Loaded Words Rated Zero)
Fig. 3. Frequency distributions of accessibility sco res
m easu red as words rated “ not disturbing” but loaded according to
complex indicators for g re a te r and le s s e r pathology groups.
48
Table 5
Rate of Im provem ent in Relation to Both M easu res of A ccessibility
(L/N+L and N um ber of Loaded W ords Rated Z e ro in D isturbance)
for Combined Hospital Groups
D egree of
Pathology
& Hospital
Im p ro v e
ment Rate
M easu res of A ccessibility
T /lvT, T Loaded W ords
L/N+L
Rated Zero
Chi
Square
E
High Low High Low
(Median = (Median = (Median =
335 days) .437) 8.9 words)
G re a te r F ast 12 5 5.18 .025
LBVAH
Slow 4 12
SFVAH
F ast
Slow
11 6
4 12
3.88 .050
(Median = (Median = (Median =
200 days) .433) 5.9 words)
L e s s e r F ast 8 8
.02 .900
SFVAH
Slow 8 7
LBVAH
Fast
Slow
6 10
8 7
1.51 .200
(Median = (Median = (Median =
335 days) .435) 8.9 words)
Combined*
G re a te r
F ast
Slow
15 9
5 15
LBVAH 4.79 .050
SFVAH
L e s s e r
F ast
Slow
14 7
7 16
SFVAH 4.37 .050
*This combined group is made up of all hospital subgroups not
significantly different from each other in the m eans of the e x p e ri
m ental variab les. See page 44.
49
It w ill be r e m e m b e re d that the Long B each l e s s e r pathology
group proved to be significantly different fro m a ll o th e r subgroups in
this study. In attem pting to d e te rm in e the relationship between L/N-i-L
and im p ro v e m e n t tim e in th is s m a ll subgroup, the F is h e r E x act
g
P ro b ab ility te s t was applied (Table 6). No significant tren d was
indicated (£ = .185).
A ccessib ility in T e r m s of N u m b er of Loaded W ords Rated Z e ro
in Relation to Im p ro v em en t T im e
The hypothesis stated that providing non-em o tio n al fac to rs a re
controlled, the n u m b e r of loaded w ords rated z ero will v a ry in v e rse ly
with im p ro v em en t r a te . Table 5 indicates that fo r the g r e a te r p athol
ogy group, the findings support the hypothesis that su b jects having
few er loaded w ords rated zero im p ro v e m o re rapidly (chi sq u are is
3.80 and p is .05). The l e s s e r pathology group again re v e a ls no sig n if
icant tre n d , for chi sq u are is 1.51 and £,>.50.
When the two g r e a te r pathology groups a r e com bined with
SFVAH l e s s e r pathology group, the relatio n sh ip is also in the predicted
d ire ctio n (chi sq u a re is 4.37 and yields a £ betw een .02 and .05).
A ccording to Siegel (54) when df = 1 and expected frequencies
a r e le ss than 5, chi sq u are m ay not be used; the F i s h e r E x act P r o b a
bility te st is suggested.
50
Table 6
Rate of Im provem ent in Relation to Both M easu res of A ccessibility
(L/N+L and N um ber of Loaded W ords Rated Z ero in D isturbance)
for Long Beach VAH L e s s e r Pathology Group
Im provem ent
Rate
M easu res of A ccessibility
T /llTlT Loaded W ords
L/N+L
Rated Z ero
High Low High Low
F i s h e r ’s
Exact
£
(Median = (Median = (Median =
335 days) .431) 5.5 words)
F a st 3 6
.1847
Slow 7 4
F ast
Slow
3
6
7
4
.1847
51
Supplementary Data
Occupational Status
Since Holmes (32) has suggested that occupational status is
related to im provem ent rate, and because this factor was not con
trolled in the present study, the data of this investigation w ere analyzed
to determ ine w hether occupational status is related to the experim ental
variables.
In Table 7 it is seen that occupational status is not significantly
related to im provem ent rate. F o r the g re a te r pathology group, chi
square is .04 and p is .80; for the le s s e r, chi square is 1.73 and p is
.2 0 .
Also neither m e a su re of accessibility is related to occupational
status (Table 8). Thus, in the case of the g re a te r pathology group,
the lack of a significant relationship between L/N-i-L and status is
ex p ressed by a chi square of .792 which yields a £ of .30; and that
between loaded words rated zero and status by a chi square of .028 and
p of .80.
In the le s s e r pathology groups, lack of a significant relatio n
ship between L/N-f-L and occupational status is revealed by a chi square
of .114 with a p of .70.
Loaded W ords Irresp ectiv e of Disturbance Rating
It has already been pointed out that in the opinion of some
52
T a b le 7
O ccu p atio n al Status in R elation to Im p ro v e m e n t Rate
fo r L e s s e r and G r e a t e r P athology G ro u p s
D e g re e
of
O ccupational
Status
Im p ro v e m e n t Rate
Chi
Square
£
P athology
F a s t Slow
G r e a t e r
N = 33
White C o lla r
M anual L ab o r
(m edian =
335 days)
9 7
9 8
.042 .80
L e s s e r
N = 31
W hite C o lla r
M anual L abor
(m edian =
200 days)
7 11
9 4
1.730 .20
53
Table 8
Occupational Status in Relation to Both M easures of Accessibility
(L/N+L and Number of Loaded Words Rated Zero in Disturbance)
for L e sse r and G reater Pathology Groups
Degree
of
Pathology
Occupational
Status
M easures of
L/N+L
High Low
Accessibility
Loaded Words
Rated Zero
High Low
Chi
Square
£
(Median = (Median =
.437) 8.9 words)
White Collar 6 10
.792 .30
Manual Labor 10 7
G reater
N =33
White Collar 8 8
.028 .80
Manual Labor 9 8
(Median = (Median =
.433) 5.9 words)
White Collar 10 8
.085 .70
Manual Labor 7 6
L e sser
N = 31
White Collar 9 9
.114 .70
Manual Labor 5 8
54
w r ite r s im p ro v em en t rate would be slowed on the b a s is of em otional
conflict r e g a rd le s s of w hether the conflicts a r e conscious o r uncon
scio u s. H ence, the total n u m b e r of loaded w ords (without re g a rd to
affective im pact) in relation to im p ro v em en t ra te was analyzed for
each of the hospital subgroups. In T able 9 it m ay be seen that the
relationship proved insignificant in e v e ry c a se . F o r each of the s u b
g roups, p w as .50.
Loaded W ords Rated D isturbing
B ec au se th e re was v ariab ility am ong the subjects with re g a rd
to n u m b e r of loaded w ords ra te d distu rb in g , it w as conceivable that
loaded w ords ra te d d istu rb in g ( j _ . n u m b e r of conflict a r e a s not
re p re s s e d ) m ig h t be related to im p ro v e m e n t tim e and to a c c e ssib ility
as defined in this study. H ow ever, Table 10 shows that for both groups
th e re is no significant relationship betw een this v a ria b le and im p ro v e
m e n t r a te . F o r the g r e a te r pathology group chi sq u a re is .27 with
p = .50; for the l e s s e r pathology group chi sq u a re equals .44 and p is
.50.
The relatio n sh ip betw een loaded w ords rated distu rb in g and
each of the two m e a s u r e s of a c c e ssib ility is not significant. T his is to
say that for the g r e a te r pathology group, w o rd s rated zero and L/N+L
give identical chi sq u a re s of 1.48 and p’s of .10 (Table 11).
Lack of significant relatio n sh ip between th ese v a ria b le s is also
55
T able 9
Rate of Im provem ent in Relation to Loaded W ords
Irresp e c tiv e of D isturbance Rating, for Hospital Groups
Combined on B asis of Sim ilar Pathology
and/or Sim ilar Means of E xp erim en tal V ariables
D egree of
Illness
& Hospital
Im p ro v e
ment Rate
Total Loaded W ords
M ore Less
Chi
Square
P
Median = Median =
G re a te r 335 days 10.9 words
LBVAH
SFVAH F ast 11 6
.69 .50
Slow 8 8
Median = Median -
L e s s e r 200 days 10.0 words
LBVAH
SFVAH F ast 6 9
.56 .50
Slow 9 7
Combined
G re a te r
LBVAH
Median =
334 days
Median =
11.2 words
SFVAH
L e s s e r
SFVAH
F ast
Slow
8 12
12 12
.44 .50
Note.— Y a te s’ correction for continuity is applied.
56
Table 10
Number of Loaded W ords Rated Disturbing
in Relation to Im provem ent Rate
for L e s s e r and G re a te r D egree of Pathology Groups
D egree of
Pathology
Loaded W ords
Rated
D isturbing
Im provem ent Rate
F a st Slow
Chi
Square
P
(median =
1.66 words)
(median =
335 days)
G re a te r
N = 33
More 7 10
0.27 .50
L ess 9 7
(median =
1.06 words)
(median =
200 days)
L e ss e r
N = 31
M ore 7 10
0.44 .50
L ess 7 7
57
Table 11
Num ber of Loaded W ords Rated Disturbing
in Relation to Both M easu res of A ccessibility
for G re a te r and L e s s e r Pathology Groups
M easu res of A ccessibility
D egree
of
Pathology
Loaded W ords
Rated
Distu rbing
L/N+L
High Low
Loaded W ords
Rated Zero
High Low
Chi
P
Square —
(median =
1.66 words)
(median =
.437)
(median =
8.9 words)
G re a te r
N = 33
M ore
L ess
More
L ess
11 6
6 10
10 6
6 11
1.48 .10
1.48 .10
(median =
1.06 words)
(median =
.433)
(median =
5.9 words)
L e s s e r
N = 31
More
L ess
More
11 7
6 7
4 10
0.22 .50
1.70 .10
L ess 10 7
58
o b se rv e d in the l e s s e r pathology group. T hus, the relationship of
w o rd s ra te d distu rb in g to L/N-t-L is e x p re ss e d by a chi sq u are of .22
and p is .50; and to w o rd s ra te d zero , by a chi sq u a re of 1,70 and a
p of .10.
C H A PTER V
Discussion
In the present study an attempt has been made to dem onstrate
that in tuberculosis when non-emotional factors a re controlled, rate of
im provem ent will tend to be directly related to accessibility of r e
pressed conflict a re a s . Two m ea su re s of accessibility were devised:
one having to do with ability to free associate to such a re a s and the
other, with the actual num ber of these a re a s.
In the case of the g re a te r pathology group, the results indicate
some confirmation of the hypothesized relationship. Each m easu re of
accessibility is associated with im provem ent rate at an acceptable level
of significance (free association to re p re sse d conflict a re a s, £ <.025;
num ber of these a re a s , £ C.05).
With the le s s e r pathology group, the findings do not suggest
confirmation of the hypothesis. Each m easure tends to be unrelated to
im provem ent rate. Since, however, the mean im provem ent tim es of
the two hospital subgroups composing the le s s e r pathology group are
significantly different, each of these groups was further analyzed.
The im provem ent rate of one of these subgroups (San Fernando
le ss e r pathology) proves to be not significantly different from either of
the g re a te r pathology subgroups. Hence, when it is combined with
them as the basis for a test of the hypothesis, the results again
60
indicate som e confirm ation of the hypothesis. Each m e a s u re of
accessibility is associated with im provem ent rate at an acceptable
level of significance. However, the level of significance is a little
lower in the case of the augmented group.
The other le s s e r pathology subgroup (Long Beach) was analyzed
separately and the findings indicate lack of confirm ation of the hypo
thesis. A clue as to why this group should not uphold the hypothesis is
found in the fact that its m ean im provem ent time is significantly less
than that of any other subgroup. This fact suggests that the initial
d isea se involvement m ay have been considerably le ss serious than for
the other subgroups. In te rm s of the psychoanalytic fram e of reference
it is possible to reaso n that when a patient has significantly less
d isea se, the amount of energy used in re p re ssin g his conflicts is
relatively unim portant as far as im provem ent rate is concerned.
P arenthetically, the two m e a s u re s of accessibility for these
combined groups (total g re a te r pathology group and SFVAH le ss e r
group) show a statistically significant degree of association. M o re
over, two n o nparam etric correlation methods were applied: Spearm an
rho was .60; Kendall tau, .45. Both indexes a re significant at the .005
level of confidence, indicating that th ere is little likelihood that the
correlation o ccu rred by chance.
That these two m e a s u re s should be related as closely as they
are seem s to lend support to the logic of th eir derivation. The
61
assu m p tio n s involved a r e that accessibility to r e p re s s e d em otional
disturbance is related to ability to free asso c ia te to r e p re s s e d conflict
a re a s; and that it is also related to the total nu m b er of such conflict
a re a s; and the m o re r e p re s s e d conflict a r e a s , the le ss the ability to
free asso ciate to them . Hence these two m e a s u re s should be related
to each o ther.
It might be worthwhile to co n sid er briefly another possible
reaso n for this co rre la tio n . Such a reason might stem from the fact
that both m e a s u re s have in com m on the selection of w ords to which
re sp o n se s a r e given only a fte r long reaction tim e. With re g a rd to
reaction tim e, it m ight be argued that while no Jung stim ulus w ords
w ere selected which had reaction tim es of le ss than two seconds, the
reaction tim e s of the w ords selected certain ly did v a ry (roughly, from
two to ten seconds). It m ight be thought that few er asso ciatio n s would
be produced by the stim ulus w ords having longer reaction tim e s , and
conversely m o re a sso ciatio n s by w ords having s h o rte r reaction tim e s.
S im ilarly, a subject whose five stim ulus w ords (used in the L/N-f-L,
m e a su re ) showed a higher m ean reaction tim e should also have a
relatively g r e a te r num ber of loaded stim ulus words rated zero in his
whole protocol and conversely, a lower m ean reaction tim e for the
five w ords should go with low er total n u m b er of loaded w ords. How
ev e r, when this hypothesis was actually tested , it was found that the
m ean s of the reaction tim es of each su b jec t’s five w ords showed no
62
significant association with either m easu re of accessibility. Since
reaction tim e seem s not to be the basis for the correlation between the
two m easu res, it appears reasonable to assum e that the m ea su re s are
related because they both m easu re accessibility.
Some comment is in o rd e r with reg ard to the em phasis in this
study on re p re sse d conflict a re a s . Such an a rea was operationally
defined as a stimulus word on the Jung Word Association T est which
c arried two complex indicators but which was rated by the subject as
being “ not disturbing.” However, according to Mahl (40), it is not
important whether emotional conflicts are unconscious or not if their
effect is to produce chronic anxiety. It should be noted, however, that
his findings a r e based on course of illness in anim als. At any rate,
as noted in the chapter on resu lts, no significant relationship was
found to exist between loaded words rated disturbing and im provem ent
rate. Also no significant relationship was found between total num ber
of loaded words irrespective of disturbance rating and im provem ent
r a t e .
These findings do not support M ahl’s hypothesis. Rather, they
suggest that rep ressed conflicts are im portant in the course of chronic
illn e s s .
To the extent that accessibility is actually m easured by the
techniques used in this study it would seem that some support is
afforded the psychoanalytic assum ptions on which the hypotheses are
63
based. It is recognized that the re su lts a re highly tentative and that
even if repeated replication produced even m o re significant re su lts, it
would be quite difficult to predict for the individual case. F u r th e r
m o re , it is recognized that tuberculosis is m ulticausal and that any one
of the many oth er factors m ay be as im portant as the psychological
facto rs. In this study th ere was no intent to find a causal relationship,
but ra th e r to seek an aspect of personality which might be of value in
helping to predict the outcome of chronic d isease and suggest for whom
psychotherapeutic intervention might be indicated.
The lim itations of the study have to do firstly with the sam ple
which is so rigidly selected that it m ay not be rep resen tativ e of m ost
tub ercu lar groups. It is re stric te d to young m ale v eteran s who have
not been exposed to the m any adverse socioeconom ic, and cultural
influences which c h a ra c te riz e many other tuberculosis patients.
Secondly, it is well known that m edical diagnoses of extent of disease
in borderline cases a re often unreliable. M oreover, the classification
used here for differentiating the two pathology groups m ay not have
been refined enough as shown in the case of the g reat difference in
im provem ent tim e of the Long Beach le s s e r pathology group. F u r th e r
m o re , there is some question as to whether the three kinds of
m edication a re strictly equivalent and actually the drugs may not have
been studied for a period of tim e sufficient to definitely determ ine
th eir relative efficacy, even though at the tim e the study was c a rrie d
64
out physicians agreed that the drugs were equivalent.
Some question might be raised with regard to having tested the
subjects at the time when they reached sem i-am b u lato ry statu s. It
might be argued that accessibility as m easu red might not have been
the sam e if the subjects had been tested early in their hospitalization.
On the other hand, as mentioned previously, it has been dem onstrated
that following an initial s tr e s s reaction, patients quickly re v e rt to
their pre-hospitalization personality. This im plies that the sam e
re p re sse d conflicts are operating and ability to free associate to them
is the sam e.
Suggested A reas for F uture R esearch
The preceding discussion has pointed up some of the lim itations
of this study. The study should be replicated with sim ila r groups of
tuberculosis patients. Also patients having this disease who a re of
different sex, age, group, and degree of emotional disturbance might
well be investigated. F u rth e r, groups with other fo rm s of pathology
(£.g. can cer patients) might be used in testing the hypothesis.
P erh ap s use of the galvanic skin reflex in conjunction with the
word association te st might resu lt in a m o re powerful and valid tech
nique. It might be, too, that a different method for locating stim ulus
w ords reflecting the patient’s conflicts is indicated. F o r exam ple, the
patient’s history could be studied in detail and words selected from this
65
history might be used instead of the Jung w ords. A variation of this
approach is suggested by D eutsch’s (21) “ associative a n a m n e sis”
which is an interviewing method based on the concept of free a s s o c ia
tion, as used in psychoanalysis. In this method, the th erap ist listens
without interrupting. He then picks out the m o st often repeated words
and incorporates them into his own sentences meanwhile observing the
patient’s reactio n s. In the approach suggested h e re , words repeated
frequently during an interview might be used as stim uli in the sam e way
that the Jung w ords w ere in this study.
It was previously conjectured that the reason one of the le s s e r
pathology groups did not substantiate the hypothesis was that a c c e s s i
bility was not im portant to im provem ent when the patients w ere
suffering from m inim al disease involvement. This might well be
investigated.
It was also theorized previously that o bsessive-com pulsive,
intellectualizing patients need not be excluded from consideration in
testing the hypotheses of this study. It might be worthwhile to test
groups of such patients if it could be shown that they will respond to
the test situation in a co-operative m anner.
The technique developed in this study might be investigated as a
possible aid in prognosis for psychotherapy, and also as a tool for
a sse ssin g psychotherapeutic p ro g re ss.
Finally, a subject for future re se a rc h might involve an
investigation of the p re se n t health sta tu s of the su b jects of th is sam p le.
A follow-up study of h o sp ital r e c o rd s , how ever, indicates that none of
the patients in the p re se n t sam ple is now hospitalized and v e r y few
have re tu rn e d at a ll since they w e re d isc h a rg e d .
C H A P T E R VI
S u m m a ry and Conclusions
While th ere is g e n e ra l feeling am ong au th o ritie s that t u b e r
culosis is m u ltic a u sa l, psychological fa c to rs a r e also co n sid e red
contrib u to ry to the on set and im p o rta n t in the c o u rse of the d is e a s e .
M edical au th o rities hold that r e s t o r co n serv atio n of energy plays a
significant p a rt in im p ro v em en t ra te . It is also held that em otional
fa c to rs in te rfe re with r e s t. H ow ever, r e s e a r c h dealing with the
possible effect of em otional fa c to rs on im p ro v e m e n t ra te is m e a g e r
and by and la rg e inconclusive.
It w as the purpose of the p re se n t study to investigate a possible
relatio n sh ip between a c c e s sib ility of r e p r e s s e d em otional d istu rb a n c e
and im p ro v e m e n t ra te in tu b e rc u lo sis when non-em o tio n al fa c to rs a re
held constant. A ccessibility w as defined in t e r m s of (a) ability to
v e rb a liz e to r e p r e s s e d conflict a r e a s , and (b) the n u m b e r of such
a r e a s . T h e se definitions of ac c e ssib ility w e re based on a ssu m p tio n s
im p licit in the psychoanalytic fra m e of re fe re n c e . Specifically, it is
a s s u m e d in psychoanalytic th erap y that the prognosis is b e tte r the
m o re the patient can fre e a sso c ia te to r e p r e s s e d a r e a s of em otional
d istu rb a n c e , and the fe w e r the a r e a s of conflict he h a s. F u r th e r m o r e
it is th e o riz e d that en erg y is ‘‘bound” in the r e p r e s s iv e p ro c e s s and is
not available for other o rg a n ism ic functions. In tu b e rc u lo sis any
68
w ithdrawal of energy is presum ed to slow the healing p ro c e ss. While
w rite rs in the field differ as to specifically how psychological factors
a re tra n slated into som atic illness, they do agree that psychotherapy
m ay alleviate physical sym ptom s to the extent that the la tte r a re con
nected with emotional factors and are re v e rsib le .
On the strength of these assum ptions, the following general
hypotheses with re g a rd to accessibility to re p re s se d em otional d is
turbance and im provem ent ra te among tuberculosis patients w ere
advanced:
1. A ccessibility m easu red by ability to free asso ciate to
re p re s se d a re a s of emotional disturbance tends to be directly related
to im provem ent rate.
2. A ccessibility m e a su re d by le s s e r num ber of r e p re s s e d
a re a s of emotional disturbance also tends to be directly related to
im provem ent rate.
Method
A g re a te r pathology group of 33 pulm onary tuberculosis male
patients and a le s s e r pathology group of 31 m ale patients ranging in
age from 20 to 40 y e a rs w ere selected. All w ere sem i-am b u lato ry and
had attained bacteriologic conversion. They w ere matched in all
factors considered relevant to the dependent variable.
The basic psychological instrum ent was the Jung Word
69
A ssociation T e st. The subject was requested to rate all the stim ulus
w o rd s on the b asis of th e ir affective im pact on him by m eans of a
ratin g scale consisting of four steps ranging from “ no” d iscom fort
(Z ero) to “ g r e a t” d iscom fort (three).
Five em otionally loaded w ords w ere then selected for each sub
je c t according to the double c rite rio n of (a) zero rating on the
affective rating scale, (b) two complex indicators.
Following this pro ced u re, the subject was asked to free
asso ciate to each of the 10 w ords for a period of one m inute.
The dependent variable was im provem ent ra te as m e a su re d in
t e r m s of n u m b er of days elapsed from the initiation of continuous
hospitalization until the attainm ent of se m i-a m b u la to ry privilege.
The independent variable was accessib ility as m e a su re d by a ratio of
the subject’s associations to loaded w ords to those to neu tral w ords,
and by the total num ber of stim ulus w ords rated not disturbing by the
subject, but which nonetheless bore com plex indicators.
F ro m these conditions it was predicted that, providing non-
em otional facto rs w ere controlled,
1. The ratio s c o re s of accessib ility would v a ry directly with
im pro v em en t rate.
2. The num ber of loaded w ords rated zero would vary
in v ersely with im provem ent ra te .
M edians of the distributions of all variables w ere found and
70
four-fold contingency tables p re p a re d . M ood’s m edian te st using chi
square w as the principal statistic employed.
Results
1. In the g r e a te r pathology group the findings tended to con
firm the hypotheses that each m e a s u re of accessib ility was related to
im p ro v em en t rate.
2. In the case of the group having l e s s e r pathology the findings
indicated no trend with relation to im p ro v em en t rate.
3. One of the l e s s e r pathology subgroups {San Fernando) did
not differ significantly fro m the g r e a te r pathology subgroups in m ean
im p ro v e m e n t ra te , accessib ility ratio, o r nu m b er of loaded words
rated zero . When these s im ila r groups w ere combined, both
m e a s u re s of accessib ility w ere significantly a sso ciated with im p ro v e
m ent r a t e .
4. The other l e s s e r pathology subgroup (Long Beach) was
analyzed and no significant relationship was found between the e x p e ri
m ental v a ria b le s. It w as pointed out that the d eg ree of illness was
undoubtedly less in this group.
5. The re su lts also indicated that n u m b er of conflict a r e a s of
which the subject m aw are, jL. e. does not r e p r e s s , is not related to
im p ro v em en t ra te .
6. A hypothesis that n u m b er of conflict a r e a s irre s p e c tiv e of
71
disturbance rating is related to im provem ent rate proved not signifi
cant.
Conclusions
The findings were guardedly in terp reted as consistent with the
psychoanalytic assum ption that physical symptomatology in so fa r a s it
is concomitant with, exacerbated by, o r contributed to, by em otional
disturbance tends to be alleviated m o re quickly among patients who
initially rev eal g re a te r ability to free associate to a re a s of re p re s s e d
disturbance o r who have few er re p re s se d conflict a r e a s to s ta r t with.
In cases of ex trem ely m inim al d isease involvement the amount
of energy used in re p re ssio n of conflicts apparently plays a relatively
unim portant part in im provem ent rate.
The im plications of the study a s an attem pt to aid in the p ro b
lem of predicting outcome of chronic disease without implying causality
was em phasized.
Future re se a rc h was suggested involving im provem ent of the
technique of this experim ent and its application to other groups of
patients. Investigation of its use as an aid in prognosis and a s s e s s
m ent of p ro g re ss in psychotherapy was also discussed.
A P P E NDIX
APPENDIX
The T horndike-L orge F requency Ratings
of Jung Word A ssociation T e st Stimulus Words
Abuse, to 22M
Angry A
Anxiety 22M
A sk, to
B eat, to
Blue
*
Big
Bird
*
Book
**
Box
*
B read
B ride 41M
B ro th er
*
C a rro t 9M
Child
# #
Choose, to A
Cold
*
Contented 12M
Cook, to
*
Cow
**
Custom A
D ance, to
*
Dead
D ear
D eportm ent 4M
D espise A
Die, to
**
Door
Expensive 28M
F all, to
* *
F alse A
F am ily
F e a r , to
F in g er
*
F low er
**
Foolish A
F rie n d
F rien d ly
**
F ro g 25M
F u r A
G lass
❖
G reen * *
Hay 46M
Head
House
H unger 37M
Ink A
K iss, to
*
Lake
*
Lam p A
Lead pencil -
Lie
* *
Long
Luck 46M
M a rry **
Money
Month
High Frequency Words:
AA 100-/MiIlion. **Among 500 m ost frequent
A 50-99/Million. w ords.
*Among 1,000 m ost frequent
w o r d s .
74
Mountain T2 Sad A
N arrow AA
Salt
Ship
Sick
Sin
Sing, to
AA
$
Needle 37M
New
Nice
*
A
#
Old Sleep, to
sfe#
Paint., to
Pam phlet
P a r t
*
9M
**
Stem
Stork
Swim, to
A
A
A
P a r t, to
**
Take ca re , to
-
P a y , to T ree
**
P ity
P lu m
A
2M
Unjust 13M
P r a y , to A Village
*
P ric k , to 15M Voyage 41M
P rid e
P u re
A
A
Wash
W ater
White
AA
**
Q u a rre l, to A
Wild
Window
*
**
Rich
❖
Women AA
Ridicule 9M Yellow
R E F E R E N C E S
REFEREN CES
1. A lexander, F ., & F re n c h , T. Studies in psychosom atic m e d ic in e .
New York: Ronald P r e s s , 1948.
2. B a rk e r, R. G., W right, B., M yerson, L., & Gonick, M. A d ju st
m en t to physical handicap and illn e s s : A survey of the social
psychology of physique and d isab ility . (Rev. Ed.) New York:
Social Science R e se a rc h Council, 1953.
3. Benjam in, J ., Colem an, J ., & Hornbein, R. A study of p e rs o n a l
ity in pulm onary tu b ercu lo sis. A m e r. J . O rthopsychiat.,
1948, 18, 704-707.
4. B e rg e r, D. The R o rsch ach as a m e a s u re of re a l-life s tr e s s . J,
c o n su lt. P sy c h o l., 1943, 17_, 355-358.
5. B e rle , B. B. Em otional facto rs and tubercu lo sis: a c ritic a l
review of the lite ra tu re . P sy c h o so m . Med., 1948, 10, 366-
373.
6. B rotm an, S. An investigation of psychological and physiological
facto rs related to ra te of re c o v e ry in pulm onary tu b ercu lo sis.
Unpublished d o cto r's d isse rta tio n , U niver. of Calif, at Los
A ngeles, 1955.
7. Calden, G. A method for evaluating the attitudes of tuberculous
patients. A m e r . R e v . T u b e r c ., 1953, 67, 722-731.
8. Calden, G., T hurston, J . R., & Lew is, W. C. The patient looks
at tu b ercu lo sis. In P . J . S p a re r (Ed.), P e rso n a lity , s tr e s s
and tu b e rc u lo s is . New York: Int. U niver. P r e s s , 1956.
P p. 275-299.
9. Cannon, W. B. The wisdom of the body. New York: W. W.
Norton, 1932.
10. C ecil, R. L., & Loeb, R. F. (Eds.) A textbook of m e d ic in e .
Philadelphia: S aunders, 1951.
11. C h aren s, S. R eg ressiv e behavior changes in the tuberculous
patient. Unpublished d o c to r's d isse rta tio n , Catholic Univer.
of A m erica, 1955.
77
12.
13.
14.
15.
16 .
17.
18.
19.
2 0 .
2 1 .
2 2 .
23.
24.
25.
C la rk e , E , R., Zahn, D. W., & H olm es, T. H. The relationship
of s t r e s s , ad ren o c o rtic a l function, and tu b erc u lo sis. A m e r .
Rev* T u b e rc ., 1954, 69, 351-369.
Coburn, F . E ditorial: E m otional problem s in the tre a tm e n t of
tu b ercu lo sis. A m e r. Rev. T u b e rc ., 1955, 71, 299-301.
Cofer, C. N., & . Shevitz, R. W ord-A ssociation as a function of
w o rd -fre q u e n c y . A m e r. J . P sychol., 1952, 65, 75-79.
Cohen, David. R o rschach s c o re s , prognosis, and co u rse of i ll
n e ss in pulm onary tu b e rc u lo s is . J . consult. P sychol., 1954,
18, 405-408.
C olem an, J . V. P sy c h ia tric contributions to c a re of tu b ercu lo sis
patients. J . A m e r . M e d . A s s ., 1947, 135, 699-702.
Davids, A. P e rso n a lity dispositions, word frequency, and word
association. J . P e r s ., 1956, 24_ , 328-338.
Day, G. Sanatorium tre a tm e n t today. J . B rit. T u b e r c . A ss.,
1955, 36, 50-54.
D eL ucia, J . J ., & Stagner, R. E m otional vs. frequency facto rs
in w ord-recognition tim e and association tim e. J . P e r s .,
1954, 22, 299-309.
D e rn e r, G. A spects of the psychology of the tu b e rc u lo u s. New
York: PaulB. H oeber, 1953.
D eutsch, F . A.pplied p sy ch o an aly sis. New York: G rune & .
Stratton, 1949.
D iagnostic stan d a rd s and classification of tu b e rc u lo s is . New
York: National T u b e rcu lo sis A ssociation, 1950.
D ittm an, A. T ., & Raush, H. L. The psychoanalytic theory of
conflict: s tru c tu re and methodology. P sy ch o l. Rev., 19 54,
6, 386-400.
D ollard, J ., & M iller, N. P e rso n a lity and p sy ch o th e rap y . New
York: M cG raw -H ill, 1950.
Dubos, R., & Dubos, J . The white plague. Boston: Little,
Brown, 1952.
78
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
Dunbar, F lan d ers. Anxiety, s tr e s s and re sp ira to ry d isea ses,
especially tu b ercu lo sis. In P . J . S p a re r (Ed.), P e rso n a lity ,
S tre ss and tu b e rc u lo sis . New York: Int. Univer. P r e s s ,
1956. P p. 208-231.
Fenichel, O. The psychoanalytic theory of n e u r o s is . New York:
W. W. Norton, 1945.
Guilford, J . P . Fundam ental statistics in psychology and educa
tion. New York: M cG raw -H ill, 1950.
H a rris , D. H. Psychological asp ects of tu berculosis. InJ, F.
G a rre tt (Ed.), Psychological asp ects of physical d isability.
Washington: F e d e ra l Security Agency, Office of Vocational
Rehabilitation Service Series N um ber 210, 1952. Pp. 97-111.
H artz, J . T uberculosis and personality conflicts. P sy c h o so m .
M ed., 1944, 6, 17-22.
H endricks, C. M. Fundam entals of tu b e rc u lo sis. Springfield,
111.: C h arles C. T hom as, 1949.
Holm es, T. H., Hawkins, N. G., C larke, E , R., Joffee, J . R.,
& Bow erm an, C. E . P sychosocial and psychophysiologic
studies of tub ercu lo sis. P sy c h o so m . M ed ., 1957, 1_9, 134-141.
Howes, D. H., & Solomon, R. L. Word frequency, personal
values, and visual duration thresholds. J. ex p . P sy c h o l.,
1952, 58, 256-270.
Hull, C. L., & Lugoff, L, S. Complex signs in diagnostic free
association. J . ex p . P sy c h o l., 1921, 4, 1-36.
Jung, C. G. Studies in word asso c ia tio n . New York: Dodd,
Mead, 1918.
Kaplan, H. I., & Kaplan, Helen S, An h isto rical survey of
psychosom atic m edicine. J . n e r v . & m e n t. P i s ., 1956, 124,
546-568.
Kendig, Isabelle V. Irre g u la r discharge and patient education.
In P . J . S p a re r (Ed.), P e rso n a lity , s tr e s s and tu b e rc u lo sis.
New York: Int. Univer. P r e s s , 1956. Pp. 459-476.
79
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
Kubie, L. S. P r a c tic a l and th e o re tic a l a sp e c ts of p sy c h o a n a ly sis.
New York: Int. U niver. P r e s s , 1950.
L ew is, W. C., L o ren z, T. H., & Calden, G. I rr e g u la r d isch a rg e
fro m tu b ercu lo sis hospitals: a m a jo r unsolved problem .
P s y c h o s o m . M e d ., 1955, 4, 276-290.
Mahl, G. F . The effect of chronic fe a r on g a stric se c re tio n .
P s y c h o s o m . M e d ., 1949, LI, 30-39.
M e rrill, B. R. Some psychosom atic a sp ects of pulm onary tu b e r
cu lo sis. J . n erv . m ent. P i s ., 1953, 117, 9-28.
M etropolitan life in su ran ce com pany s ta tis tic a l bulletin, Nov.,
1952.
Mood, A. M. Introduction to the theory of s ta t i s t i c s . New York:
M cG raw -H ill, 1950.
M oorm an, L. J . The psychology of the tuberculous patient. A m e r .
Rev. T u b e r c ., 1948, 57, 528.
M oran, L. J ., F a irw e a th e r, G. W,, F is h e r , S., & M orton, R. B.
P sy ch o lo g ical concom itants to rate of re c o v e ry fro m tu b e r c
u lo sis. J . c o n su lt. P s y c h o l., 1956, 20, 199-203.
M oran, L. J ., F a irw e a th e r, G. W., & M orton, R. B. Some
d e te rm in a n ts of su c c e ssfu l and u n su ccessfu l adaptation to
ho spital tre a tm e n t of tu b e r c u lo s is . J . consult. P sy c h o l.,
1956, 20, 125-131.
Muldoon, J . F . Some psychological concom itants of tu b ercu lo sis
and hospitalization: a p re lim in a ry study. P sy c h o so m .M ed .,
1957, 19, 307-314.
Newman, J ., & S p a re r, P . J . V erbal b ehavior and o th e r c o r
r e la te s of adjustm ent to hospitalization am ong tu b e rc u lo sis
patients. In P . J . S p a re r (Ed.) P e r s o n a lity , s t r e s s and
tu b e rc u lo s is . New York: Int. U niver. P r e s s , 1956. Pp. 450-
458.
P au leen , M. Some relatio n sh ip s betw een personality and beh av
io r in hospitalized tu b ercu lo sis patients. A m e r . R e v ,
tu b erc. pulm. D is., 1957, 76, 2-10.
00
50. P o stm a n , L., & Schneider, B. H. P e r s o n a l values, v isu al r e c o g
nition, and re c a ll. P s y c h o l. R e v ., 1951, _58, 271-284.
51. R ebner, L. Role of perso n ality as a facto r in pulm onary
tu b e rc u lo sis. Unpublished d o c to r’s d iss e rta tio n , U niver. of
M ontreal, 1956. Cited by E . D. W ittkow er in P . J . S p a re r
(Ed.), P e r s o n a lity , s tr e s s and tu b e r c u lo s is . New York: Int.
U niver. P r e s s , 1956, P . 161.
52. S ala z a r, J ., & T h urston, J . (D iscussion) I r r e g u la r d isc h a rg e , its
c au ses and suggested c u re s; and An ap proach to the ir r e g u la r
d isc h a rg e problem . N e w sle tte r for psychologists in t u b e r c .,
1955, 2, 37-38.
53. Selye, H. The s t r e s s of life. New York: M cG raw -H ill, 1956.
54. Siegel, S. N o n p a ra m e tric s ta tis tic s for the b eh av io ral s c ie n c e s .
New York: M cG raw -H ill, 1956.
55. S p a re r, P . J . (Ed.) P e r s o n a lity , s tr e s s and tu b e r c u lo s is . New
York: Int. U niver. P r e s s , 1956,
56. S tre e te r, G. A. P h a n ta sie s of tu b e rc u lo sis patients. P s y c h o s o m .
M e d ., 1957, 3_9, 287-292.
57. T horndike, E . L., & L o rg e, I. The t e a c h e r ’s word book of
30,000 w o r d s . New York: C olum bia U n iv e r . P r e s s , 1944.
58. T h u rsto n , J . R., & Calden, G. Intelligence fa c to rs in ir r e g u la r
d isc h a rg e am ong tu b e rc u lo sis patients. J . c o n s u lt. P s y c h o l.,
1954, 18, 404-408.
59. V e rn ie r, C la ire M., W hiting, J . F ., & M e tz e r, M. L.
D ifferential prediction of a specific b ehavior fro m th re e p r o
jective te c h n iq u e s. J . c o n s u lt. P s y c h o l., 1955, 19, 175-182.
60. V e rn ie r, C la ire M. P re d ic ta b ility in tre a tm e n t of tu b ercu lo sis
patients. In P . J . S p a re r (Ed.) P e r s o n a lity , s t r e s s and
tu b e rc u lo sis , New York: Int. Univer. P r e s s , 1956. Pp. 344-
352.
61. W eiss, E ., & E nglish, O. S. P s y c h o s o m . M e d . (2nd Ed.).
Philadelphia: S au n d ers, 1949.
81
62.
63.
64.
65.
6 6 .
67.
6 8 .
White j R. W. The abnorm al p e rso n ality . New York: Ronald
P r e s s , 1948.
W ittkower, E . D. Psychological aspects of pulm onary tu b e rc u
losis. In P . J . S parer (Ed.) P e rso n a lity , s tr e s s and
tuberculosis. New York: Int. Univer. P r e s s , 1956. Pp. 153-
174.
W ittkower, E. D., D urost, H. B., & Laing, W. P . P sy c h o
som atic study of the course of pulmonary tub ercu lo sis. A m e r.
Rev. T u b erc., 1955, 71, 201-219.
W ittkower, E. D. A psychiatrist looks at tu b e rc u lo sis . London:
National A ssociation for the Prevention of T uberculosis, 1949.
Wolff, H. G. S tre ss and d is e a s e . Springfield, 111.: C h arles C.
Thom as, 1953.
Woodworth, R., & Schlosberg, R. E xperim ental psychology. New
York: Henry Holt, 1954.
Z im m e r, H. Self-acceptance and its relation to conflict. J,
co n su lt. P sy c h o l., 1954, 18, 447-449.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Delinquency As A Function Of Intrafamily Relationships

PDF

Prediction Of Therapeutic And Intellectual Potential In Mentally Retardedchildren

PDF

A Semantic Differential Investigation Of Critical Factors Related To Achievement And Underachievement Of High School Students

PDF

An Investigation Into The Communication Style Of Suicidal Individuals

PDF

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

An Experimental Study Of Aggression Habit Patterns And Sex Role

PDF

Measuring Thought Process As An Ego Function In Schizophrenic, Mentally Retarded And Normal Adolescents By Means Of The Rorschach

PDF

Perceptual Defense And Somatization: A Comparison Of The Perceptual Thresholds Of Obese And Peptic Ulcer Patients

PDF

Acculturation And Value Change

PDF

Age Differences In Serial Reaction Time As A Function Of Stimulus Complexity Under Conditions Of Noise And Muscular Tension

PDF

Role Expectations Of American Undergraduate College Women In A Western Coeducational Institution

PDF

Sex-Role Preferences Of Early Adolescents In Relation To Adjustment

PDF

First Steps In An Attempt To Construct An Objective Test Of Character Structure

PDF

Anxiety Level And The Repression-Sensitization Dimension In Desensitization Therapies

PDF

The Effects Of Feedback On The Communication Of Medical Prescription To Diabetic Patients

PDF

Perception Of The Power Structure By Social Class In A California Community

PDF

Attitudes Of Ministers And Lay Leaders Of The American Baptist Conventionof The State Of Washington On Selected Social Issues

PDF

Self-Worth, Future Goals, Mood And Time Perception

PDF

Psychological Variables In The Mother Related To Infant Feeding Patterns

PDF

Effects Of Different Treatment Procedures On Reading Ability And Anxiety Level In Children With Learning Difficulties

Asset Metadata

Creator Rosenthal, Harry J. (author)

Core Title Improvement Rate In Tuberculosis As A Function Of Accessibility To Repressed Emotional Disturbance

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, general

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Seward, Georgene H. (committee chair), Garwood, Victor P. (committee member), Jacobs, Alfred (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-27890

Unique identifier UC11357714

Identifier 5902614.pdf (filename),usctheses-c18-27890 (legacy record id)

Legacy Identifier 5902614-0.pdf

Dmrecord 27890

Document Type Dissertation

Rights Rosenthal, Harry J.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA