Indian J. Psychiat.

(1974), 16, 283-286

THE MIDDLESEX HOSPITAL QUESTIONNAIRE (M. H. Q.)

STANDARDISATION ON A HINDI VERSION
Downloaded from http://journals.lww.com/indianjpsychiatry by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4

O. N. SRIVASTAVA, M.D., M.R.C. Psych. D.P.M. 1 and V. K. BHAT, M.B.B.S. 2

XMi0hCywCX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdgGj2MwlZLeI= on 04/09/2023

(Received 26th September 1973)

Introduction: 3. The correlation coefficients between
The Middlesex Hospital Questionnaire the subtests were moderate to low,
is a short, clinical diagnostic self-rating showing that they measure different
scale for psychoneurotic patients, cons- aspects of neurosis but at the same
tructed by Crown and Crisp (1966) . The time, all correlations were positive,
authors' aim was a rapid quantification of showing that there is some general
common symptoms and traits relevant to underlying factor (i.e.; neuroticism)
the conventional diagnostic categories of affecting all the tests. Hence the
neurotic illness — to provide a rapid ap- sum total of all the subscales can be
proximation to what would be expected taken as the neuroticism score.
from a diagnostic psychiatric interview.
Since its construction, the M.H.Q.
Instead of simply telling that somebody
was used by various authors, all of whom
is more ill than others, the authors claim
demonstrated its utility and its superiority
that the M.H.Q. gives a "quantitative
over other scales like M . P . I , and P.E.N.
clinical profile " as the test consists of six
(Cockett, 1969 ; Mc Kerracher et al, 1968;
subscales having 8 questions each. The
Young et al, 1971 ; Wolkind and Forrest,
subscales a r e :
1972) . Regarding the Hysteria subscale,
1. Free-floating anxiety (FFA) Cockett argued that it measured extra-
2. Obsestional traits and symptoms (OBS) version rather than hysteria. Moreover,
3. Phobic anxiety (PHO) Young et al, (1971) found that the Hysteria
subscale of the M.H.Q. correlates with
4. Somatic concomitants of anxiety (SOM)
the E scale of the P . E . N , test whereas
5. Neurotic depression (DEP) the rest of the subscales and the F . F . A .
6. Hysterical personality traits (HYS) in particular, correlate positively with the
N Scale of the P.E.N, test. It also seems
The authors validated the test on the that the HYS subscale measures the
following criteria: hysterical personality traits and not the
1. Each subtest differentiated the neu- hysterical neurosis.
rotics from the normals at a highly
significant level. Further evaluation of the M.H.Q. was
done by Crown et al, (1970) who ad-
2. The subtest scores correlated with ministered the test to 1208 adult males
the clinical ratings of the particular and provided extensive standardisation
diagnostic category. figures. They came to the conclusion that

1 Reader & Head of the Department of

Institute of Medical Sciences,
Psychiatry.
Banaras Hindu University,
2 Clinical Registrar, Department of
Varanasi—5.
Psychiatry
284 O. N. Srivastava and V. K. Bhat

age and social class could be ignored in The hysteria scale, however, was
normal clinical use and sex affected the grossly modified. It was in fact, constructed
subtests in a varying manner. Crisp and anew, taking into consideration the usual
Priest (1971) investigated a general prac- clinical presentation of hysteria in India.
tice sample of 778 middle-aged persons and By going through t h e past records of
plolted their psychoneurotic profiles ac- the psychiatric clinic attached to the
cording to age, sex etc. They found that Department of Psychiatry, an inventory
except in the HYS scale (in which males of symptoms complained of by the hys-
and females scored equally), women scored terical patients was m a d e and the most
significantly higher in all the subscales. common symptoms w e r e included in the
scale. Only one item (item N o . 18 of
P r a b h u (1972) used the M . H . Q . the original scale) was retained as such.
(Original English form) in the Indian The number of items w a s limited to 8, so
setting and found it to be of same that the total number of items in the test
efficacy as claimed by the original authors. remained 48 and all the subtests had a
He also found the repeat reliability of uniform structure and scoring system.
each of the subtests to be good and the The order of the items is also changed
total score of the test to correlate + 0 . 6 2 in the Hindi Version. The 8 items of
with the N score of the E . P . I . each subscale have been p u t into 2 clusters
of 4 items each. The first cluster of 4
Aims and objectives : items from each of the subscales follow
i) To prepare a Hindi Version of the in a sequence and then the second cluster
Middlesex Hospital Question- of each of the subscales follows in the
naire. next sequence. The first half of the
ii) To obtain preliminary standardi- whole test thus contains the first halves
sation figures in a normal popula- of all the subtests. The order in which
tion. the clusters of items appear i s :
iii) To establish the reliability and
FFA, OBS, PHO, SOM, DEP and HYS.
validity of the Hindi Version.
(The questionnaire and the scoring key
Materia? and Methods : may be had from the authors on request).
A preliminary form of the Hindi Ver- T^e Hindi Version so developed was
sion was prepared by translating the administered to a group of 120 normal
original 43 items of the M . H . Q . into Hindi. peop'e (80 males -f- 40 females) . This
Communication of the real meaning was was a heterogeneous group consisting of
given more emphasis than making a word University Students, the hospital staff,
to wo'd translation. Copies of this Hindi student nurses and some middle aged
Version were circulated, along with the men from different walks of life. The
original English test among the staff of the t c , ; was also administered to the 1st Year
Department of Psychiatry. The staff Medical Students of the Institute of
c o m p r i s e d two psychiatric consultants, Medical Sciences (56 males) and this
six Post-graduate students and a Psy- groun was a homogeneous group in which
chiatric Social Worker. They were asked all the subjects were of the same age-
to go t'nrouf'.h all the items critically and group, sex (all males), socio-economic and
sug'ios' improvements. All such sugges- professional status. The second group was
tions were incorporated in the final form deliberately chosen as w e wanted to see
of the version. whether the standardisation figures obtain-
 M.H.Q. — Hindi Version 285

ed from the first group are stable No attempt was made to see the repeat
enough. The first year Medical Students reliability of the test as it was thought
form a definite homogenous group which that the neurotic symptoms have in any
can be followed up at least for five years. case little temporal stability in a clinical
The table 1 shows that the norms are situation.
quite stable as the scores of the two
' normal' populations are very similar. VALIDITY :
When the scores for the 40 females were
analysed separately, they were found to To test the validity, the scale was
be simliar to those of males in all the sub- administered to 100 neurotics (75 males
scales except the PHO Scale where they + 25 females), who attended the psychia-
scored significantly higher than males. tric clinic for treatment during the period

TABLE 1
Standardisation figures for the M H.Q. (Hindi Version)
Scale N o r m a l s — Heterogeneous g r o u p . Normals — H o m o g e n e o u s group
(N = 120; 80 males + 40 females) (N = 66 male medical s t u d e n t s )
Mean S. D. Mean. S. D.
FFA 45 3.0 4.7 3.6
OBS 7.8 2.7 7.6 3.2
PHO* 4.4 3.0 3.6 2.7
SOM 4.7 3.1 4.9 3.0
DEP 5.3 3.0 5.6 2.8
HYS 4.2 2.1 4.3 2.7
TOTAL S C O R E 30 9 11.3 30.7 13.4
• I n t h e PHO scale, females snore significantly higher than males. Moans a n d S t a n d a r d D e v i a t i o n s
for females and males are 5.8 + 2.8 and 3.5 + 2.8 respectively (P<0.001).
NOTE: Differences between t h « scores of t h e homogeneous and heterogeneous g r o u p s a r e
insignificant. The P H O score of t h e Medical S t u d e n t s ' g r o u p is lower beoause it
contained only males.

RELIABILITY : between 1st May and 31st August, 1973.

Reliability was calculated by the split- The ability to read and understand the
half method from the scores of the Medical questionnaire was the only criterion for
selection of the subjects. The group
Students Group. The split-half reliability
largely consisted of anxiety neurotics,
of the whole test (i.e. between the first
depressives and hysterics. Some referred
24 items Vs. the next 24 items) was 0.70
and undiagnosed (Psychosomatic and
(by spearman brown formula) . For the
' functional') cases were also included in
subscales the reliability coefficients were this group.
as follows :
Subscale Reliability It was found that the M . H . Q . scores
FFA oTsi differentiate the neurotics from the
OBS 0.76 normals at a statistically highly signi-
PHO 0.5* ficant level. The P value is much less
SOM 0.68 than 0.001 for all the subscales. (Table 2).
DEP 0.74 The unpaired t-test was used to find the
HYS 0 66
significance of difference.
288 O. N. Srivastava and V. K. Bhat

TABLE 2
Norms for the middlesex hospital questionnaire (Hindi Version) Normals vs Neurotics
Scale NlOrmnls (N = 120) Neurotics (N = ,00) * t Value P value
Moan S D. Mean S.D. less t h a n
FFA 4.5 3.0 10.1 3.6 12.2 0.001
OBS 7.8 2.7 9.8 3.3 4.9 0.001
PHO 4.4 3.0 7.3 3.5 6.6 0.001
SOM 4.7 3.1 10.3 3.8 12.0 0.001
DEP 6.3 3.0 9.7 3.7 9.6 0.001
HYS 4.2 21 8.2 3.0 11.4 0.001
T O T A L SCORE 30.9 11.3 55.4 15.2 13.3 0.001
* T h e r vallue for t h e signifioanoe cif 0.001 is 3.34 ( d f - 218).

Conclusion: Cris P, A . H . a n d P r i e s t , (1971) : ]Psychoneurc

Profiles in middle-age. Brit. J. Psychiat. 119,
It is evident from this study that the 335.
Hindi Version of the M.H.Q. is a very
sensitive, reliable and valid instrument for Crown, S. and Crisp, A . H. (1966) : A short
differentiating the neurotics from the clinicrl diagnostic self rating scale for
normals. It is multidimensional, covering psychoneurotics — the M. H . Q. Brit. J.
various aspects of neurosis but at the same Psychiat. 112, 917.
time relatively short and simple, taking
hardly 7—10 minutes for administration. Crown. S., Duncan, K . P . and Howell, R. W.
(1970) : Further Evaluation of the M . H . Q .
Summary: Brit. J. Psychiat. H6, 33.

A Hindi Version of the Middlesex Ivlc Kerracher. D. W., Loughnane, T. and Watson,
Hospital Questionnaire was prepared and R. A. (1968) : Self-mutilation in female
preliminary standardisation was done in psychopaths. Brit. J. Psychiat, 114, 829.
an Indian Population. Its reliability and
validity were established. Prabhu. G. G. (1972) : Clinical utility of the
Middlesex Hospital Questionnaire in India.
Indian J. Psychiat. 14. 127.
Acknowledgement:
The authors are grateful to all the mem- Wolkind and Forrest, (1972) : Low back pain:
bers of the staff of the Department of Psychiatry. A psychiatric investigation. Postgr. Med. J.
Institute of Medical Sciences, for their co-opera- 48, 76.
tion and help in the conduct of this work.
Young, J. P. R., Fenton, G. W. and Lader, M.
(1971) : The inheritance of neurotic traits:
REFERENCES:— A Twin Study of the Middlesex Hospital
Cockett. R . U 9 6 6 ) : A short diagnostic self- Questionnaire. Brit. J. Psychiat. 113, 393.
rating scale in the pre^adult Remand setting.
Brit. J. Psychiat. 115, 1141.