Patient-Uniformity Myth 311subgroups have been identi“ed, few attempts have been
made to evaluate the differential ef“cacy of treatments cus-
tomized to patient subgroup characteristics.
As chronic pain involves both physical and psychosocial-
behavioral factors, we have argued that it might be most
appropriate to consider a dual-diagnosisfor chronic pain pa-
tients„one based on physical mechanisms and a second on
patients• unique set of psychological and behavioral features
(Turk, 1990). The results of several studies conducted by
Turk and Rudy (1988, 1990) and con“rmed by other investi-
gators in different countries demonstrate that psychological
assessment data that incorporate cognitive, affective, and be-
havioral information can be integrated using empirical taxo-
metric methods. These data and methods can serve as the
basis for a classi“cation system of chronic pain patients along
relevant dimensions or axes„physical-symptomatic and
psychosocial-behavioral.
Using empirical methods, Turk and Rudy (1988, 1990)
have been able to identify group patients based on their:
1.Reports of pain severity and suffering.
2.Perceptions of how pain interferes with their lives, includ-
ing interference with family and marital functioning,
work, and social and recreational activities.
3.Dissatisfaction with present levels of functioning in fam-
ily, marriage, work, and social life.
4.Appraisals of support received from signi“cant others.
5.Life control incorporating perceived ability to solve prob-
lems and feelings of personal mastery and competence.
6.Affective distress including depressed mood, irritability,
and tension.
7.Activity levels.
They identi“ed three distinct patient pro“les:1.Dysfunctional (DYS) patients, who perceived the severity
of their pain to be high, reported that pain interfered with
much of their lives, reported a higher degree of psycho-
logical distress due to pain, and reported low levels of
activity.
2.Interpersonally Distressed (ID) patients, with a common
perception that signi“cant others were not very supportive
of their plight.
3.Adaptive Copers (AC) patients, who reported high levels
of social support, relatively low levels of pain and per-
ceived interference, and relatively high levels of activity
(Turk & Rudy, 1988).
The classi“cation system has been replicated in several
studies, with different pain syndromes (Turk, Okifuji, Sinclair,
& Starz, 1998; Turk & Rudy, 1990), and in different coun-
tries. Turk and colleagues (1990; Turk et al., 1998) found that
the psychosocial-behavioral subgroups are independent on
demographic, disease status, and physical pathology.
The results alluded to above support the recommendation
for a dual-diagnostic approach. Further the data suggest that,
although different physical diagnostic groups may require
common biomedical treatments, targeting the pathophysio-
logical mechanisms underlying each disorder; they may also
bene“t from speci“c psychosocial interventions tailored to
their psychosocial-behavioral characteristics.
As a preliminary step to evaluate the differential treatment
response by patient subgroups, Turk and colleagues (Turk,
Rudy, Kubinski, Zaki, & Greco, 1996; Turk et al., 1998)
examined the outcomes of “bromyalgia syndrome and tem-
poromandibular pain dysfunction syndrome patients to com-
mon treatments. Although as a group, both diagnostic sets of
patients responded to common treatments, examination of
the treatment outcomes by the psychosocial-behavioral sub-
groups revealed that these subgroups did not each have the
same response. The results of these studies provide data sup-
porting the existence of different psychosocial-behavioral
subgroups of pain patients and the potential for treatment
matching. Moreover, the outcomes suggest that treating all
patients the same may dilute the ef“cacy of various treat-
ments. The same logic and methodology that we have used to
identify psychological subgroups can be adopted to identify
physical-symptom-based subgroups. Speci“c interventions
may have their greatest utility when matched to both dimen-
sions„physical and psychological with particular subgroups
of patients.
To date, there have been few attempts to customize treat-
ment so as to match them to patient characteristics. Clinical
investigations should, therefore, be conducted to determine
the relative utility of different treatment modalities based on
the match of treatment to patient characteristics, and to pre-
dict which patients are most likely to bene“t from what com-
bination of therapeutic modalities. Rather than accepting the
pain-patient homogeneity myth,the “eld might be advanced
by asking: •What treatment, provided by whom, in what way,
is most effective for which patients, with what speci“c prob-
lem, and under which set of circumstances?Ž
The identi“cation of subgroups, regardless of the methods
used, does not mean that the resulting classi“cation will in-
corporate all important features of the patients. Subgroups
should be viewed as prototypes, with signi“cant room for in-
dividual variability with a subgroup. Thus, matching treat-
ment to subgroup characteristics will also need to consider
and address unique characteristics of the individual patient.
The subgroup customization should “t somewhere between