Académique Documents
Professionnel Documents
Culture Documents
This study determined the prevalence ofpersonality disorders among clients in treatmentfor addiction; this prevalence was compared with thosefound in similar studies and in clinical samples of individuals sufferingfrom other
Axis I disorders. Our sample comprised 255 subjects. The first edition of the Millon Clinical Multiaxial Inventory
(MCMI) was used. Only 11.8% of the subjects did not score over 84 on any of the 11 Axis II scales. Over one-half
had a score of 84 or higher on the passive-aggressive and dependent-personality scales. The mean number of
scales in the 84+ category was 2.68. Comparisons show that this sample was more severe in most cases.
(Can J Psychiatry 1999;44:592-596)
Key Words:personality disorder, substance disorder, Millon Clinical Multiaxial Inventory, treatment, gender
Objectives
This study determines the prevalence of PDs among individuals admitted to public treatment centres in Quebec and compares these results with findings of similar studies and with
those of clinical samples suffering from other Axis I disorders.
Method
Our sample comprised 255 clients182 men and 73
womenin 8 SD treatment centres in Quebec. The subjects
were francophone and aged 18 years or over. The mean age
was 34.5 years: 12% aged 18-24 years; 41% aged 25-34
years; 34% aged 35-44 years; and 13% aged 45 years or
older. Testing took place after 7 days of treatment in order to
exclude individuals suffering from acute withdrawal. Participation was voluntary, and all subjects signed a consent form.
We compared our results with those of 5 studies (4,8,19-21)
that had used the MCMI-I (Table 1).
To better understand the specificity of SD, our results were
also compared with those of 3 clinical samples from Quebec.
The first comprised 180 subjects treated for erectile or orgasmic disorders. The other 2 samples included individuals
treated for sexual impulse problems at a medicolegal clinic:
one comprised 44 men who had committed a rape or had intrusive rape fantasies, and in the other were 87 men who had
committed pedophilic acts. All subjects were francophone.
All subjects signed a consent form.
We used the first edition of the MCMI (1), a self-report inventory asking 175 true or false questions. Our results focused on
the 11 personality scales (PSs) described in Table 2. The
MCMI was translated into French (22). Based on
August 1999
Diagnosis at
admission
Type of
treatment
Time of testing
Ness and
others (20)
30
Opiate abuse
or dependence
Brown (19)
50
All substance
dependencies
12-step
treatment from
21-30 days
detoxification
After 21-30
days of
treatment
Craig and
others (4)
86
Opiate abuse
or dependence
Detoxification
After
detoxification
Craig and
others(4)
107
Cocaine abuse
Detoxification
After
detoxiHcation
Marsh and
others (8)
159
Craig and
others (21)
Craig and
others (21)
Axis 11
No score +
Schizoid
Avoidant
After 48-72
hours of
abstinence
593
MCMI validity index (1 or more on the Y scale); 9 were rejected because of exaggeration of symptoms (sum of scales 1
to 8 greater than 164); 6 were rejected because of missing
data; 3 were rejected because the subjects were under age 18
years. In the other clinical samples, 25 tests (8%) were rejected.
Statistical tests designed to compare means and proportions
were used. The types of tests used are specified below. Differences between genders were calculated when the number of
subjects allowed it.
Results
The first column of Table 2 shows the results by scale. Subjects are classified in this category for each scale on which
their score is 84+. Only 11.8% did not score over 84 on any
scale, over one-half scored on the passive-aggressive (56.9%)
106
Alcohol abuse
and dependent (52.9%) PS, and 0% scored on the compulsive
or dependence
PS. For the 74 cutoff point, the third column of Table 2 shows
Opiate abuse
100
98.8% of subjects to be above this threshold. Over 50%
or dependence
scored on either the passiveaggressive, dependent, avoidant, borTable 2. Millon Clinical Multiaxial Inventory base rate score!
derline, or schizoid PS. None scored
By scale
Highest score
Average score
more that 74 on the compulsive PS.
Base rate score > 84
Base rate score > 74
Significantly more women than men
scored 84+ on the histrionic, schizo%
Rank
%
Rank
%
Rank
Rank
%
typal,
borderline, and paranoid PS.
11.8
1.2
0
11.8
Methadone
treatment
After
admission to
treatment
Dependent
52.9
69.4
21.2
75.85
Histrionic
10.6
24.7
2.7
58.63
Narcissistic
12.5
20.8
10
4.3
56.58
10
Antisocial
12.5
26.3
3.5
60.90
Compulsive
0.0
11
0.4
0.0
11
39.66
11
Passive-aggressive
56.9
76.5
20.8
82.30
Schizotypal
7.1
10
22.0
0.4
10
65.85
Borderline
21.2
55.3
3.9
75.45
Paranoid
12.5
31.4
4.7
69.05
Mixed
9.0
psychometric qualities (that is, internal consistency, testretest reliability, and discriminant validity), this translation is
reliable and valid (23). Millon's taxonomy parallels that of
the Diagnostic and Statistical Manual of Mental Disorders
(DSM-III).
Base rate (BR) scores were computed for each scale. A BR
score of 74 or more (74+) means that the subject shows signs
of the disorder; a score of 84 or more (84+) signals prominence of a disorder. Fifteen tests were rejected because of the
594
Scale
Female
Total
Schizoid
70.37 (23.44)
71.19(25.59)
70.61 (24.03)
Avoidant
81.04(21.11)
81.10(23.99)
81.05(21.92)
Dependent
75.34 (24.66)
77.15(25.83)
75.85 (24.97)
Histrionic
59.82(19.23)
55.64 (28.70)
58.63(22.38)
Narcissistic
56.95(21.29)
55.66(24.92)
56.58(22.35)
Antisocial
61.68(22.10)
58.96(21.99)
60.90(22.06)
Compulsive
39.18(18.02)
40.88(21.85)
39.66(19.16)
Passive-aggressive
82.67 (23.36)
81.38(25.30)
82.30(23.89)
Schizotypal
65.60(11.16)
66.48(14.14)
65.85(12.07)
Borderline*
73.57(12.19)
80.12(16.40)
75.45(13.81)
Paranoid
68.11 (12.96)
71.40(16.31)
69.05(14.05)
'P<Q.O\.
Vol 44, No 6
Table 4. Comparison of Millon Clinical Multiaxial Inventory scale score between 4 samples
Alcohol and drug
abuse
Sexual
dysfunction
Rape
Pedophilia
Total
(n = 255)
(n = 173)
(n = 34)
(n = 79)
(n = 541)
Schizoid
70.61
45.95-
61.50
61.77-
60.86
Avoidant
81.05
53.01-
70.50-
72.62-
70.19
Dependent
75.85
57.40-
67.15
76.38
69.48
Histrionic
58.63
55.08-
56.15
52.65
56.46
57.53
59.12
Axis II
Narcissistic
56.58
63.35+
60.41
Antisocial
60.90
66.55+
67.38
59.56
62.92
Compulsive
39.66
60.14+
51.29+
51.63+
48.69
Passive-aggressive
82.30
51.65-
63.94-
63.76-
68.64
Schizotypal
65.85
58.21-
62.09
65.29
63.09
Borderline
75.45
62.6&-
67.12-
69.96-
Paranoid
69.05
69.46
67.32
+This sample mean is significantly higher than the drug alcohol abuse sample.
-This sample mean is significantly lower than the drug alcohol abuse sample.
August 1999
an annex because of doubts as to its specificity. Millon excluded this scale from the MCMI-III.
Clinical Implications
Results according to gender show that PDs are more prevalent among women, no matter how the results are analyzed.
Numerous studies have demonstrated similar results (30).
Women, however, have as good a prognosis as men and, in
some cases, a better one (31). The specific effect of PD on
treatment outcome in reference to gender, however, is unknown.
Limitations
Acknowledgements
This research was supported by a grant from the Conseil quebecois
de la recherche sociale to the team Recherche et Intervention sur les
Substances psychoactives - Quebec (RISQ).
17. Nace EP, Davis CW. Treatment outcome in substance-abusing patients with a
18.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
595
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
596
Vol 44, No 6
Resume
Cette etude a determine la prevalence des troubles de personnalite parmi les clients en traitement pour toxicomanie ; cette prevalence a ete comparee a celles constatees dans des etudes semblables et dans des echantillons
cliniques de personnes souffrant d 'autres troubles de I 'axe I. Notre echantillon se composait de 255 sujets. La premiere edition de I 'Inventaire clinique multiaxial de Millon (MCMI) a ete utilisee. Seulement 11,8% des sujets n 'ont
pas obtenu de resultat superieur a 84 a aucune des II echelles de I'axe II. Le nombre moyen d'echelles dans la
categorie des 84 etplus etait 2,68. Les comparaisons indiquent que cet echantillon etait plus gravement malade
dans la plupart des cas.