588
P. Enthoven et al.
intervention (see intervention description below) had
greater enablement than patients in the control group
(13, 14). This shows that enablement can mirror the
results differently from traditionally used outcome
measures. To our knowledge, no studies have investi-
gated the measurement properties of the PEI for this
group of patients.
The aim of the study was to investigate content va-
lidity, construct validity, internal consistency, and self-
rated change (SRC) for the PEI in patients with chronic
musculoskeletal pain treated in different settings. Our
hypotheses were that the PEI is unidimensional, and
that it has fair to moderate relationships with health-
related measures, such as disability, mental and general
health, and self-reported work ability.
MATERIALS AND METHODS
Design
The study included participants from 3 studies of chronic
musculoskeletal disorders: a whiplash-associated disorders
(WAD) group, a cervical radiculopathy (CR) group, and a mixed
chronic pain (MixCP) group. For the WAD and the CR studies
these were secondary analyses of multicentre, prospective, ran-
domized controlled clinical trials (Clinical Trials.gov, number
NCT01528579 and NCT01547611, respectively) (15, 16). For
the MixCP study this was a prospective pragmatic cohort study.
All 3 groups completed questionnaires at baseline. The WAD
group completed follow-up questionnaires after a 3-month reha-
bilitation programme, the CR group 3 months after surgery, and
the MixCP group after a 6–8 weeks rehabilitation programme.
Participants and settings
Whiplash-associated disorders group. Participants were recrui-
ted from primary care in 6 counties between February 2011 and
May 2012. Participants were aged between 18 and 63 years, with
whiplash injury grade 2 or 3 in the preceding 6–36 months, and
had received neck-specific treatment. Detailed eligibility criteria
have been published previously (15).
Cervical radiculopathy group. Participants were recruited from
3 spinal centres and were scheduled for surgery between January
2010 and December 2014. Participants were aged between 18
and 70 years with persistent CR symptoms and verified disc
disease for at least 2 months on magnetic resonance imaging
(MRI). Detailed eligibility criteria have been published pre-
viously (16).
Mixed chronic pain group. The study included consecutive
patients aged 18–65 years with a referral for multimodal reha-
bilitation (MMR) between August 2010 and December 2012.
All participants had disabling non-malignant chronic pain with
a duration of at least 3 months that was due to musculoskeletal
disorders (MSD; according to the ICD-10 (International Clas-
sification of Diseases) plus the potential for an active change
in their lives. Participants were recruited from 6 healthcare
centres in a County Council region in south-east Sweden. All
participants were offered MMR by an interdisciplinary team.
Exclusion criteria for participants in all 3 study groups were
severe psychiatric disorders, neurological diseases, drug abuse,
or insufficient competence in the Swedish language.
www.medicaljournals.se/jrm
Intervention
Whiplash-associated disorders group. For the current study
participants who received neck-specific exercises with (NSEB)
or without (NSE) a behavioural approach (pain management
strategies and goal setting) were included (15, 17). Both groups
received information about neck anatomy and the purpose of the
exercises. The interventions lasted for 12 weeks and included
exercise guided by a physiotherapist twice a week plus home
exercises.
Cervical radiculopathy group. For the current study, the surgical
procedure included anterior cervical decompression and fusion
(ACDF) (14). Initial postoperative care at the spinal centre in-
cluded advice about posture and ergonomics, information about
movements and tasks to avoid, and mobility exercises for the
shoulders. At 6 weeks the participants had a routine visit to the
surgeon and received instructions about mobility exercises for
the neck from a physiotherapist (14).
Mixed chronic pain group. The treatment consisted of an MMR
programme based on a biopsychosocial model that considers
the individual’s somatic, psychological, environmental, and
personal characteristics. MMR included, for example, pain edu-
cation, physical exercise and cognitive behavioural therapy. The
interdisciplinary rehabilitation team included different health-
care professions, e.g. an occupational therapist, physiotherapist,
physician, psychologist, and social worker. Patient interaction
in goal-setting was encouraged. The duration of the programme
was 6–8 weeks. The treatment comprised group-based sessions
complemented by individual treatment sessions or counselling,
depending on the patient’s needs. The sessions were held once
or twice a week (range 7–17).
Assessment
The participant-reported questionnaires cover important out-
come domains for the evaluation of chronic pain clinical trials,
as recommended by the Initiative on Methods, Measurement,
and Pain Assessment in Clinical Trials (IMMPACT) (18), and
are reported in Table I.
Patient enablement: main outcome in the current study. After
the treatment period, the PEI was used to measure the patient’s
perceived change in ability to understand and cope with his or
her health issues (4). A minor change was made to the intro-
ductory statement of the original PEI to make it more relevant
to the specific study. Specifically, “As a result of your visit to
the doctor today, do you feel you are …” was changed to “As
a result of the treatment for your problem(s), do you feel you
are …” followed by the original 6 items (see Table V). The PEI
points for much better/much more is 2; better/more is 1; and
same, less, or not applicable is 0. Thus, the total score is between
0 and 12 (4), and was calculated only for those participants
who answered all 6 questions. A higher score indicates higher
enablement. The validity and reliability of the PEI have been
found to be acceptable (3, 4, 6, 10). The Swedish version of
the PEI shows high internal consistency and moderate to good
reliability for a single visit in primary care (19).
Perception of received care. Global perceived effect (GPE) was
measured with the question “Compared with before treatment,
how would you describe your complaint/problem now …?”.
For the WAD and CR studies the answer was rated on a 6-point
scale that ranged from “complete recovery” to “much worse,”
and for the MixCP study on a 7-point scale that ranged from
“very much improved” to “very much worsened”.