Bendix (2000)

138 patients with chronic low back pain

Inclusion criteria:

-be in a precarious work situation

Exclusion criteria:

-herniated nucleus pulposus

-surgically treatable instability

-inflammatory back disease

-osteoporosis

-cancer

-medical or psychiatric condition preventing intensive physical training

-already received a disability pension

Intervention (n=64):

The functional restoration program focussed on self-responsibility and activity with a multidisciplinary approach. The program lasted for 3 weeks, 9h/d.

The daily schedule included:

-physical training under supervision of a physical therapist (2h)

-occupational therapy under supervision of an occupational therapist (1.5h)

-group therapy sessions led by a psychologist (1.5h)

-stretching supervised by a physical therapist (0.5h)

-theory on anatomy, physiology and pathology taught by a physician, physical therapist, occupational therapist or social worker. (1h)

-recreational activities under supervision of the physical therapist and/or occupational therapist (1h)

Control (n=74):

Patients in the outpatient intensive physical training program received the same physical training as patients in the intervention group. The control program lasted for 8 weeks, three times per week, 1,5 h/session.

Self-efficacy/disability (ADL assessment; median [0-30])
Base:
I:16 (12–20) / C: 16 (12–21)
12mo:

I: 12 (IQR=6-21) / C: 13 (IQR=9-19)
p-value= 0.41

Back Pain intensity (median, (IQR) [0-10])
Base: I: 5.1 (4–7) / C: 6.0 (5–7)
12mo: I: 5.1 (2-7) / C: 5.7 (3-7.3)

Work participation (days sick leave; median)
Base: -

12mo: I: 5.5 (0-113) / C: 2.5 (0-301)

P-value = 1.00

Loss to follow-up:

1 year I: 17.2% / C: 31.1%

Jousset (2004)

86 French patients with low back pain

Inclusion criteria:

- LBP patients, age 18-50 years

- Engaged in a nonlimited work contract

- risk of inability to work by LBP

- Threatened in their job situation by CLBP

- Not relieved by conventional medical or surgical interventions

Excluding criteria:

- LBP of specific origin (< 4 months)

- Recent spinal surgery

- Cardiac or respiratory abnormalities after exercise stress test

- Precluding psychiatric disorders

- Receiving of disability pensions

Intervention (n=44):

Functional Restoration Program ( FRP) for 5 weeks

5 days per week

6 hours per day

Aspects:

Intensive physical training

-warm-up, stretching, proprioception

-strengthening exercises (physiotherapist -supervised)

-aerobic activities

-endurance training

-balneotherapy

Occupational therapy providing work simulation

-individual interventions such as consultations with the physiatrist, psychologist and dietician

Control (n=42):

Active individual therapy (AIT)

5 weeks

3x per week (+ 2x p/w 50 min.)

1 hour per day

Aspects:

Active exercises

Starting with 2 weeks with focus on flexibility, range of motion, pain coping strategies, followed by strengthening exercise and functional training and advice to perform regular sports activities

Disability (ADL-Dallas questionnaire component ADL [0-100])
Base: I: 53.7 (16.7) / C: 50.3 (16.7)

6mo I: 36.7 (23) / C: 41.5 (24.4)

Between-group difference p=0.36

Disability (Quebec Disability Scale[0-100])

Base: I: 34.6 (15.4) / C: 31.6 (15.9)
6mo I: 22.0 (16) / C 22.9 (17.7)

Between-group difference: p=0.80

Pain intensity (during the past 48 hours was evaluated on a 10-cm horizontal visual analogical scale [0-10])
Base: I: 5.0 (2.2) / C: 4.6 (2.2)

6 mo I: 3.1 (2.5) / C: 4 (2.8)

Between-group difference: p=0.16

Anxiety/depression (HAD scale [0-21])
Base: I: 17.0 (6.5) / C: 14.3 (6.1)

6mo I: 12.7 (7.2) / C: 13.4 (6.4)

Between-group difference: p=0.72

Work participation (number of sick-leave days)
Base: -

6mo I: 28.7 (44.6) / C: 48.3 (66.0)

Between-group difference: p=0.12

Overall loss to follow-up: 1%

Kääpä (2006)

120 women (Finland) with low back pain

Inclusion criteria:

- age 22-57 yo

Excluding criteria:

-Clinical symptoms suggesting an acute disc prolapse accompanied by nerve root entrapment.

-recent back surgery

-severe cardiovascular or other disorder interfering with active rehabilitation

-specific back disorder

-severe mental illness

-more than 90 days off work because of LBP during preceding year.

-pension within 2 years

-pregnancy

-ongoing/planned LBP rehabilitation

Intervention (n=59)

Multidisciplinary rehabilitation program(MR) for 8 weeks (total of 70h),

Aspects:

-cognitive-behavioral stress management

-applied relaxation sessions

- back school education including occupational Intervention

-physical exercise program.

- in small supervised groups (6-8)

Control (n=61)

Individual physiotherapy (IP) for 6-8 weeks consisting of 10x1h sessions

Aspects:

-passive pain treatment

-light active exercise

- recommendation of general physical training

-light home-exercise program

Disability (Oswestry disability index)

Base I: 25.4 (10.6) / C: 23.8 (11.7)

After re I: 20.9 (10.1) / C: 21.6 (11.4)

6 mo I: 20.4 (11.6) / C: 18.0 (11.5)

12 mo I: 18.9 (12.8) / C: 18.5 (12.4)

24 mo I: 19.7 (14.3) / C: 19.3 (13.1)

Not statistically significant

Low back pain intensity [0-10]

Base: I: 4.6 (1.9) / C: 5.0 (2.6)

After re I: 3.3 (2.5) / C: 3.4 (2.4)

6 mo  I: 3.3 (2.5) / C 3.4 (2.5)

12 mo I: 3.6 (2.7) / C: 3.4 (2.5)

24 mo I: 3.5 (2.6)/ C: 4.0 (2.9)
Not statistically significant

Depression (DEPS [0-30])

Base I: 7.5 (5.2) / C: 6.7 (5.5)

After re I: 5.5 (5.5) / C: 5.7 (5.2)

6 mo I: 5.7 (4.6) / C: 5.8 (5.7)

12 mo I: 6.6 (5.8) / C: 5.0 (4.0)

24 mo I 6.7 (5.3) / C: 5.7 (4.7)
Not statistically significant

Work participation (number of sick leave days during last year)
Base: o

after 1y. (p=0.65)

0 days I: 61.5%/ C: 56.4%

1-30 days I: 32.7% / C: 34.6%

>30 days I: 5.8% / C: 9.0%

Loss to follow-up

6mo I: 2% / C: 7%

12mo I: 10% / C: 11%

24mo I: 17% / C: 25%

All participants were women

At inclusion, participants had only mild to moderate symptoms à this could (partially) explain the absence of differences between the groups

Sick leaves probably not reliable outcome since the temporary workers were afraid to go on sick leave during working hours

The intervention contrast between the IP and MR groups was, however, somewhat decreased by the instructions of exercise

treatment

Some subjects, particularly those doing

physically hard work, perceived participation during working

hours in MR program as burdensome

Roche (2007, 2011)

132 French patients with chronic low back pain

Inclusion criteria:

- Aged 18-50 years,

- Diagnosed with LBP for ≥3mo,

- Patients on sick leave or at risk of work disability.

Exclusion criteria:

- Patients with malignant,

traumatic, infectious, or inflammatory LBP,

- Patients with acute LBP, spondylolisthesis, or cardiac/ respiratory insufficiency,

- Patients with articular or neurologic impairment,

- Patients with psychiatric disorders,

- Patients receiving disability pensions or refusing to participate in the study.

Intervention (n=68): functional restoration program (FRP). For 5wks, 6h a day, 5d/week in groups of 6-8 patients. Exercises were supervised by a physiotherapist. Each week patients attended a clinic with the specialist in physical medicine and rehabilitation who was the medical supervisor of the program. They were referred to a psychologist at least once in the first week and for further treatment if requested.

Control (n=64): individual therapy program (AIT). For 5wks and included individual rehabilitation with a private physiotherapist for 1h 3x/week and individual exercises to be performed at home for 50min 2x/week. Exercises were supervised directly by a physiotherapist.

Both groups (n=132):

- Patients were off work during 5wks,

- At T0 and T5, patients were assessed in a rehabilitation centre by physiatrists.

- Physical evaluations were performed by physiotherapists.

Measurements: 5wk, 12mo

Disability (Daily activities - subscale Dallas Pain Questionnaire, [all subscales 0-100%])
Base: FRP: 51.9% / AIT: 51.0%

T0-5 weeks: FRP: -21.5% / AIT: -17.2%

T0-12 mo: FRP: -20.3% / AIT: -10.4%
Not statistically significant

Disability (Work and leisure – subscale Dallas Pain Questionnaire, [all subscales 0-100%])

Base: FRP: 51.9% / AIT: 58%

T0-5 wks: FRP: -22.0% / AIT: -19.8%

T0-12 mo: FRP: -22.8% / AIT: -17.7%
Not statistically significant

Disability (Social interaction – subscale Dallas Pain Questionnaire, [all subscales 0-100%])

Base: FRP: 30.7% / AIT: 27.4%
T0-5 weeks: FRP: -13.6% / AIT: -4.1%

T0-12 mo FRP: -15.7% / AIT: -1.4%
Statistically significant

Pain (VAS [0-10])

Base: FRP: 4.7 (2.1) / AIP: 4.5 (2.1)

T0-5wk FRP: -1.9 / AIP: -1.5

T0-12mo: FRP: -1.7 / AIP: -1.0
Not statistically significant

Anxiety and depression (subscale Dallas Pain Questionnaire, [all subscales 0-100%])
Base: FRP: 36.9% / AIP: 30.8%

T0-5wk: FRP: -17.6% / AIT: -7.4%

T-0-12 mo: FRP: -15.6% / AIT: -4.8%
Statistically significant

Return to work (%)
Base: -

T0-5 weeks: FRP: 86.8% / AIT: 85.7%

p-value>0.05

- Compliance with both programs was not assessed.

- The main limitation of this study is the lost to follow-up rate. At 5wks some patients did not complete all measurements (available FRP: all data n=63, endurance n=63, return to work n=67, DPQ n=67. Available AIT: all data n=56, endurance n=56, return to work n=6, DPQ n=63). After 1 year, 19 (14.4%) patients were missing and they were more frequently included in AIT than in FRP (AIT: 15/64 patients, FRP: 4/68 patients, p=0.006).

- Intention-to-treat analysis could not be performed (Roche 2011).

Ronzi (2017)

105 French patients with chronic low back pain

Inclusion criteria:

Participants were subjects of working age suffering from non-specific chronic LBP according to the usually accepted scientific criteria.

Patients could be included if they were aged 18 to 55, were able to sign the informed consent form, suffered from a non-specific LBP for at least three

months without improvement which had led to at least one month’s sick leave during the preceding year and/or three months’ sick leave during the preceding two years, and were on an open-ended

or a fixed-term contract of work in the public or private sector.

Exclusion criteria:

No possibility to follow one of three programs of study

Pregnant women, new mothers or mothers who are breastfeeding.

No affiliation to the French social system.

Intervention group: Functional Restoration Program (FRP) (n=49): The group performed exercises supervised by a physiotherapist who adjusted the exercise intensity

to each participant every week. The endurance training was adapted to each participant’s heart rate and to the exercise stress test performed before the program. This program involved six hours of treatment a day, five days a week in group of six to eight patients.

Control group: Ambulatory Individual Physiotherapy (AIP) (n=54)

Ambulatory Individual Physiotherapy (AIP) included individual rehabilitation with a private physiotherapist for one hour, three times a week and individual exercises to be performed at home twice a week for 50 minutes. The mixed strategy includes AIP combined with five one-day group sessions.

Measurements: at inclusion (I: n=49 / C: n=54) and 12-months follow-up (I: n=43 / C: n=31)

HRQoL SF-36 PCS (0-100), median (IQR)

Pre: I: 35.7 (29.4 to 39.5)/ C: 35.6 (31.9 to 37.2)

12mo: I: 39.1 (33.8 to 50.4) / C: 37.5 (33.0 to 46.8)

HRQoL SF-36 MCS (0-100), median (IQR)

Pre: I: 43.3 (32.1 to 49.8) / C: 41.2 (36.1 to 50.8)

12mo: I: 48.3 (42.1 to 53.4) / C: 48.9 (41.4 to 54.8)

Dallas Pain Questionnaire daily activity subscale, % (IQR)

Pre: I: 63 (51-72) / C: 54 (48-69)

12mo: I: 51 (12 to 64.5) / C: 54 (36 to 63)

Dallas Pain Questionnaire work and leisure subscale, % (IQR)

Pre: I: 65 (50 to 75) / C: 56 (50 to 75)

12mo: I: 45 (10 to 65) / C: 40 (25 to 65.5)

Dallas Pain Questionnaire anxiety/depression subscale, % (IQR)

Pre: I: 45 (20 to 60) / C: 40 (15 to 55)

12mo: I: 30 (5 to 45) / C: 30 (5 to 45)

Dallas Pain Questionnaire social interaction subscale, % (IQR)

Pre: I: 30 (20 to 40)/ C: 25 (10-45)

12mo: I: 15 (0 to 42.5) / C: 25 (15 to 50)

Intensity of pain on VAS (mm) (0-100), median (IQR)

Pre: I: 54 (45 to 65) / C: 45 (24 to 65)

12mo: I: 45 (25 to 59) / C: 33 (19 to 48)

Hospital anxiety and depression scale (HADS) (0-21), median (IQR)

Pre: I: 17 (12 to 21) / C: 16 (12 to 21)

12mo: I: 11.5 (7.5 to 18) / C: 13 (8 to 19)

Number of sick-leave days during the

preceding 12 months, median (IQR)

Pre: I: 256 (195 to 325) / C: 219 (126-318)

12mo: I: 50.5 (0 to 200) / C: 45 (0-98)

Only the total HADS score is reported, not separately for the subscales anxiety and depression

Tavafian (2011, 2014, 2017)

132 Iranian patients with chronic low back pain

Inclusion criteria:

- CLBP patients aged ≥18 years.

- Pain for ≥90 days.

- Primary causes of CLBP were herniated intervertebral disc, facet arthropathy, and lumbar arthrosis.

Exclusion criteria:

- Back surgery within last 2 years.

- Recent vertebral fracture.

- Vertebral malignancy.

- Infection in the back.

- Spondylolisthesis.

- Spinal stenosis.

- Inability to participate in the multidisciplinary program sessions,

- Current pregnancy.

- Residency outside of Tehran, insufficient address / phone number for follow ups.

- Difficulty understanding Farsi language.

- Unwillingness to enter to the study or comply with the study protocols.

Intervention group (n=97):

5x 2-hour sessions in one week (group), plus monthly booster sessions and telephone counselling. Aspects:

- Education (anatomy, risk factors, lifestyle advice, posture, diagnosis, pain education).

- Exercises (stretching, strength, relaxation exercises).

- Psychologist (coping, stress, perceptions of control, emotional reactions, problem solving, relaxation).

- CBT (maladaptive cognitions, fear avoidance, activity participation,

adjustment to pain).

- Motivational counselling.

- Medication (analgesics, muscle relaxants, NSAIDs, anti-depressants)

Control group (n=100):

Usual care, medical management, mostly medication prescription.

Both groups were evaluated every 3 months by the same rheumatologist.

Throughout the study, medications were prescribed for patients in the 2 groups as needed.

6mo: I: 97 vs C: 100

12mo: I: 87 vs C: 91

24mo: I: 82 vs C: 83

HRQoL (SF-36), two example domains:

SF-36 Physical function
Base: I: 54.61 (23.27) / C: 54.53 (23.30)

3mo I: 68.64 (23.39) / C: 60.93 (22.04)

6mo I: 77.77 (18.71) / C: 63.69 (21.88)

12mo I: 80.3 (18.6) / C: 64.4 (22.8)

24mo I: 56.2 (22.5) / C: 55.6 (22.8)

SF-36: Mental Health
Base: I: 47.43 (13.96) / C: 44.00 (13.10)
3mo: : I: 66. 2 (20.4)/ C: 58.3 (23.1)
6mo: I: 67.2 (22.1)/C: 60.9 (23.4)

12 mo: I: 71.8 (20.2)/ C: 58.9 (24.9)
24 mo: I: 48.1 (13.6)/C: 43.8 (12.01)

Time*Group t0-t6: p-value <0.05
7 subscales including physical function, role physical, role emotional, bodily pain, general health, social function, and vitality

Time*Group t3-t12: p-value <0.05 physical function, mental Health

Time*group t3-t24: p-value <0.05 physical function, mental health

Disability (Quebec Disability Scale)
Base: I: 35.45 (20.19) / C: 33.08 (19.69)

3mo I: 23.48 (18.54) / C: 32.70 (18.19)

6mo I: 18.65 (16.14) / C: 27.19 (17.85)

12mo I: 17.4 (16.4) / C: 24.4 (18.3)

24mo I: 35.5 (18.9) / C: 33.5 (18.9)
Time*Group t0-t6:p-value <0.01

Time*Group t3-t12: p-value=0.9

Time*group t3-t24: p-value=0.9

Disability (Ronald-Morris Disability)
Base: I: 9.80 (5.07) / C: 10.04 (5.28)

3mo I: 9.01 (5.71) / C: 10.56 (5.78)

6mo I: 7.03 (5.49) / C: 8.80 (5.68)

12mo I: 6.01 (5.8) / C: 8.9 (6.6)

24mo I: 9.9 (4.9) / C: 9.8 (5.3)

Time*Group t0-t6:p-value=0.01

Time*Group t3-t12: p-value=0.5

Time*group t3-t24: p-value=0.4

Loss to follow-up

Analyzed at 12mo I: 87; C: 91
Analyzed at 24mo I: 82; C: 83

- Limited generalizability: majority of studied participants were women.

- There is no exact data regarding the types and quantities of prescribed drugs patients consumed during the study period.

- Authors did not ask participants what other health service resources, or

whether additional visits were used during the extended follow-up.

- One physician who was not fully

masked to the intervention, treated the patients.

Castel (2013)

Inclusion criteria:
- female sex
- a diagnosis of FM based on the diagnostic

criteria of the American College of Rheumatology
- age

between 18 and 60 years
- between 3 and 8 years of

schooling.

Exclusion criteria:

severe chronic pain pathology (e.g., sciatica or complex regional pain syndrome), having been diagnosed with inflammatory rheumatic disease, being physically unable to perform the exercises, an open wound, a skin disease,

being under psychiatric and/or psychological treatment

within the past 3 years, significant suicidal ideation, cognitive or sensorial deterioration that impedes an adequate

follow up to the treatment, or a pending legal resolution for disability.

Intervention (n=81):
CBT delivered by trained clinical psychologist

 Physical therapy delivered by a trained physiotherapist.


Control (n=74):
Conventional pharmaceutical treatment, prescribed by physician.

Disability (FIQ total score)
Base: I 64.6 (16.0); C 66.6 (17.4)
Post: I 47.7 (20.2)‡; C 65.9 (16.1)
3mo: I 55.5 (19.3)§; C 64.6 (17.6)
6mo: I 55.8 (20.9)§; C 67.8 (18.4)

12mo: I 58.8 (20.5); C 69.6 (17.2)
Statistical significance: post, 3mo, 6mo

Pain intensity (VAS 0-10)
Base: I 6.8 (1.4); C 7.1 (1.6)
Post: I 5.7 (1.9); C 6.9 (1.8)
3mo: 6.4 (1.9); C 6.8 (1.8)
6mo: 6.4 (1.9); C 7.0 (1.9)

12mo: 6.7 (1.6); C 7.1 (1.8)
Statistical significance: post

Anxiety/depression (HADS)
Base: I 21.9 (8.0); C 23.2 (8.1)
Post: I 14.3 (9.0)‡; C 21.7 (8.4)
3mo: I 15.2 (9.1)‡; C 20.6 (8.5)
6mo: I 16.2 (9.3)‡; C 21.5 (8.5)

12mo: 17.1 (9.9)†; C 22.8 (9.2)

Statistical significance: post, 3mo, 6mo, 12mo

Van Eijk-Hustings (2013)

203 patients with fibromyalgia

Inclusion criteria:
Patient with fibromyalgia (according to the American College of Rheumatology criteria) recruited from outpatient rheumatology clinics
- 18 – 65 years

Exclusion criteria:
Only for MD group:
- pregnancy
- involvement in litigation concerning work disability procedures
- use of other non-pharmacological treatments such as psychological or physical treatment, interfering with the intervention
- alcohol or drugs abuse
- use of walking devices.

Intervention group (n=108)Multidisciplinary intervention (MD):
Phase 1 consisted of a 12-week course, three half days per week, with two therapy sessions of 1.5 h duration per day. A trained and experienced multidisciplinary team offered a program of sociotherapy, physiotherapy, psychotherapy and creative arts therapy, using not only group interaction as an additional tool but also paying attention to the patient’s specific needs.
Sociotherapy included education and connected the parts of the program.

Physiotherapy was given twice a week. The program was focused on graded activity, based on time-contingent instead of pain-contingent training and aimed to improve physical fitness and functioning and at learning to enjoy exercise.

Psychotherapy was given once a week and consisted of general information about FM and pain mechanisms. Methods of core qualities [19], rational emotive therapy [20–22] and transactional analysis [17, 18] were used in the sessions.

Creative arts therapy was given once a week and focussed on the opportunity to express feelings by visual arts instead of verbal expressions.

Phase 2 was an aftercare program and consisted of five meetings, scheduled over a period of 9 months. The purpose of these meetings was to repeat the key messages about coping in order to preserve the behavioural change achieved in phase 1

Usual care (UC) (n=48)
Received care as usual that comprised at least individualised education about FM and lifestyle advice by a rheumatologist or a specialised rheumatology nurse within one or two consultations, but could also include a diversity of other treatments such as physiotherapy or social support from the rheumatology nurse

Reported results are Estimated marginal means *±standard error.

EQ5D (-0.59-1)

Effect size between I and C at 18 mo:

0.12 (−0.22, 0.46)

EQ-VAS (0-100)

Effect size between I and C at 18 mo:

0.22 (−0.12, 0.56)

Fibromyalgia Impact Questionnaire-total (FIQ)

Effect size between I and C at 18 mo: 0.25 (−0.09, 0.59)

* Cohen defined an ES of ≥0.2 as small, ≥0.5 as moderate and ≥0.8 as large. ES calculations were based on the results of the mixed-model analyses and were considered as statistically significant if zero was not included in the CI.

Martín (2012, 2014a, 2014b)

132 Spanish patients with fibromyalgia.

Inclusion criteria:

- Patients diagnosed with FM.

- Widespread pain for at least 3mo in combination with pain on palpation in 11-18 specified tender point sites.

- Aged >18 years.

- Continuous chronic pain for at least 6mo.

Exclusion criteria:

- Decline participation in study.

- Suffering from a severe psychiatric (psychosis or suicide risk) or organic disorder.

- Were involved in employment-related legal proceedings related to their FM.

PSYMEPHY group (n=54):Patients received the same

pharmacologic treatment as the control group. They also participated in 6wk of PSYMEPHY delivered by a team (incl. physician, clinical psychologist, physiotherapist).

Patients were divided into groups (12p) and received biweekly group sessions (a total of 12 sessions). Various sessions e.g.:

- psychologist + educational activities with a physician and psychologist.

- psychologist + physiotherapist.

Various CBT components (cognitive,

physiological, behavioural, educational) were addressed.

Control group (n=56): Current care (=pharmacological treatment tricyclic antidepressant, an analgesic, and an opioid central analgesic.

Fibromyalgia Impact Questionnaire (FIQ):total score [0-100]

Base: I: 76.28 (13.57) / C: 76.23 (14.88)

6mo: I: 70.33 (16.48) / C: 76.81 ± 14.18

p-value=0.02

T0-t6mo: ES: 0.45 (95%CI 0.07 to 0.83, p=0.02)

HADS: Anxiety
Base: I: 13.83 (3.39) / 13.39 (3.45)

6mo:I: 13.41 ± 4.31 / C: 12.75 ± 4.55
p-value=0.08

ES: -0.07 (95%CI -0.44 to 0.31, p=0.72)

HADS: Depression
Base: I: 10.63 (4.51) / C: 10.57 (4.06)

6mo: I: 9.77 ± 4.09 / C: 10.2 ± 4.22
p-value=0.31

ES: 0.17 (95%CI -0.21 to 0.54, p=0.19)

ES: effect size (0.20 = small effect; 0.50 = medium effect); positive ES favors the EG; negative ES favors the CG

- Only PSYMEPHY group had a follow-up period of 12mo. Comparisons were not made between PSYMEPHY and control groups at the 12-month follow-up.

- Limited generalizability:
1) participants were selected from patients referred to a hospital pain management unit (sample might experience greater impact of FM on HRQOL than patients treated in primary care),
2) females were over-represented in our sample.

Skouen (2006)

126 patients with fibromyalgia and chronic widespread pain. Pain was

present on digital palpation in at least 11 of 18 tender

points according to the American College of Rheumatology criteria. Participants had to have a job with no

fixed ending date and be sick-listed more than 50%.

Intervention (n=42)
The extensive multidisciplinary treatment program lasted for 4 weeks, with 6 h of sessions 5 days per week. The program included group sessions, education, exercise training and occasional workplace intervention. The group sessions discussed cognitive coping strategies and gave advice (2 hours per week). Anti-depressant medication was used if necessary. Patients were encouraged to use analgesics as little as

possible. Education sessions involved anatomy, pain mechanism, exercise training and applying mental coping strategies at work and daily life (2 hours per week). The extensive program also focused on avoiding fear. The patients received an individually tailored graded

exercise training program based on physical tests (Ljunggren, 1993). The patients exercised daily under continuous supervision by physiotherapists to the patients for

1.5–3.5 hours per day.

Control (n=85)
Treatment as usual group. After the screening and examination by a physician, this group was referred to their general practitioner. In

Norway, some general practitioners give CWP patients pain medication, anti-depressant medication and refer

them to physiotherapists or chiropractors. Other general

practitioners have started to change their way of handling these patients. More patients receive a cognitive

 treatment approach and traditional physiotherapy is changed to body awareness treatment. Data showed that 50% of the patients in the reference group had seen a physiotherapist and 20% had tried alternative treatment of different kinds. This percentage did not differ significantly from the two other groups.

Return to work
In men: Percentage return to work seem similar in the intervention group and in the control group, also not significantly significant.

In women: Until ±24 months, patients in the intervention group returned to work more often. AT 54 months of follow-up, the difference in number of days absent from work between the intervention and control is 206.95 days.

Becker (2000)

114 patients with chronic non-malignant pain patients

Inclusion criteria:
- age above 18 years

-no psychiatric diseases present

-no illegal use of opioids

-all relevant medical or surgical investigations treatments completed prior to referral

Exclusion criteria:
-Patients who were unable to fill in questionnaires were not included

Intervention group (n=56):
The MPT-group received outpatient treatment at the Multidisciplinary Pain Centre. The treatment programme was carried out on an individual basis and when considered necessary planned and initiated after a multidisciplinary evaluation. The treatment was of primarily cognitive-behavioural nature and included one or several of the following components: (1) education on the physiology and psychology of pain; (2) teaching of pain management strategies (e.g. relaxation training); (3) analgesic treatment; (4) socio-economic counselling and (5) physiotherapy. The physiotherapy focused primarily on teaching of exercise programs and education in biomechanics. The staff consisted of anaesthesiologists, psychologists, physiotherapists, nurses and a social worker. Throughout the treatment period all patients were seen regularly by the pain specialists. Between consultations patients received telephone counselling from nurses at the pain centre.

Control group (n=58):

In the GP-group the intervention consisted of a single initial consultation during which the pain specialist evaluated the patient together with the GP in his/her consultation room. The pain specialist made a medical record and a pain analysis. Based on this the pain specialist and the GP established a treatment plan. Treatment consisted primarily of analgesic tailoring, however, during the consultation the importance of education on the physiology and psychology of pain and behavioural pain management strategies were also emphasized. The protocol allowed patients access to additional monospeciality treatment if needed.

HRQoL (SF-36)

For different SF-domains; only social functioning at 6 months showed a clinically relevant difference between intervention and control. This was however not significantly different (p-value>0.05).

Pain intensity (VAS)
Base: I: 67 (15) / 65 (23)
3 mo: I: 52 (24)/ C: 63 (20)
6 mo: 52 (24)/ C: 65 (25)
p-value 6 mo: <0.05

HADS-anxiety
Base: I 7.8 (4.0) / C: 7.8 (4.2)
3 mo: I: 7.2 (4.2)/ C: 7.5 (4.3)
6 mo: I: 7.1 (4.5))/ C: 8.2 (4.0)
p-value 6 mo: >0.05

HADS-depression
Base: I: 6.8 (4.9) / C: 6.4 (4.2)
3 mo: I: 6.3 (5.2)/ C: 6.4 (4.4)

6 mo: I: 6.1 (5.4)/ C: 6.6 (4.3)
p-value 6 mo: >0.05

Turner-Stokes (2003)

126 patients with chronic pain

Inclusion criteria:

-Age ≥ 18 yo

-Patients experienced pain for > 6mo

-Conventional treatment failed

-No need for hospital transport to clinic

-No major changes in medical management anticipated for next 6 months

Exclusion criteria:

-referral for further treatment or investigation

Intervention (n=66):

Group-based multidisciplinary program for 8 weeks

1 afternoon per week

aspects:

-relaxation strategies

-cognitive coping strategies

-encouragement for exercise

-pacing of daily activities

With support from psychology, physiotherapy, occupational therapy, medical staff

- in small groups (8-10)

control (n=47)

individual program

8 weeks

1h /2wk’s

Aspects:

Tailored advice regarding physical activity and exercise delivered by the physiotherapist (via the psychologist)

Measurements: 2mo, 6mo, 12mo

Pain severity (WHYMPI, mean diff)

Base: I: 4.26 (1.02) / C: 4.0 (1.14)

2mo -0.07 (95%CI -0.50 to 0.35, p=0.73)

12mo -0.17 (95%CI -0.59 to 0.24, p=0.41)

Anxiety (STAI, mean diff)

Base: not reported
 2mo -2.3 (95%CI -6.21 to 1.59, p=0.24)

12mo -3.43 (95%CI -7.81 to 0.94, p=0.12)

Depression (BDI, mean diff [0-63])

Base: not reported
2mo 0.91 (95%CI-1.89 to 3.71, p=0.52)

12mo -2.19 (95%CI -4.69 to 0.32, p=0.09)

The programs were anticipated to be of approximately similar cost when group-based therapy was delivered in groups of 10 people.

Loss to follow-up:

12mo I: 33% / C: 34%

Loss to follow-up = 33% (at 12mo) especially by patients with spinal pain due to travelling time.

Pain duration difference between groups at time of inclusion

Study reference

(first author, publication year)

Was the allocation sequence adequately generated? ^a

Definitely yes

Probably yes

Probably no

Definitely no

Was the allocation adequately concealed?^b

Definitely yes

Probably yes

Probably no

Definitely no

Blinding: Was knowledge of the allocated

interventions adequately prevented?^c

Were patients blinded?

Were healthcare providers blinded?

Were data collectors blinded?

Were outcome assessors blinded?

Were data analysts blinded?

Definitely yes

Probably yes

Probably no

Definitely no

Was loss to follow-up (missing outcome data) infrequent?^d

Definitely yes

Probably yes

Probably no

Definitely no

Are reports of the study free of selective outcome reporting?^e

Definitely yes

Probably yes

Probably no

Definitely no

Was the study apparently free of other problems that could put it at a risk of bias?^f

Definitely yes

Probably yes

Probably no

Definitely no

Overall risk of bias

If applicable/necessary, per outcome measure^g

LOW

Some concerns

HIGH

Bendix (2000)

Probably yes;

Reason: randomization, or rather stratification by minimization, was intended to equalize age, gender, days of sick leave, pain, disability, physical measures and smoking

No information;

Probably no;

Reason: patients had to accept the result of randomization and knew assignment to the treatment group.

Blinding of assessor successful for approx. 80% of the patients.

Probably yes;

Reason:

Loss to follow-up:

1 year I: 17.2% / C: 31.1%

Several ITT analyses done. Baseline values of drop-outs were compared. Tendency towards worse outcomes among drop-out patients can be noted, but similar in both arms.

Also no bias found between patients who did/did not show up for follow-up assessment

Definitely yes;

Reason: at posttreatment, all relevant outcomes were reported

Study doesn’t include reflection on methodology

Little to no description of instruments for outcome measurements

Some concerns

Baseline imbalance on work capability

Jousset (2004)

Definitely yes;

Reason: Block randomization was performed using an eight element

permutation table

No information;

Probably no;

Reason: The

evaluation was not blind, but cooperation at all times between all evaluators and standardization tended to reduce possible

discrepancies. Blind allocation to I or C for patients not reported

Probably yes;

Reason: no dropout at 6 months among the patients having completed the program. Small infrequent dropout during follow-up assessment

Definitely yes;

Reason: All relevant outcomes were reported;

Probably no;

Reason: the evaluation at 6 months could not be performed based on all patients. Four patients could not come to the clinic and did not perform the physical evaluation.

Some patients refused part of the tests, either because of pain or fear of pain.

The evaluation was not blind, but cooperation at all times between

all evaluators and standardization tended to reduce possible

discrepancies.

Several baseline differences were significant (such as HAD scale)

Some patients with lower evaluations at inclusion did not perform the tests at 6 months, and this has possibly induced an overestimation of the results, probably more important in the

AIT group.

Some concerns

Anxiety/depression (sign baseline diff)

PILE-test (selective/MNAR loss to follow-up)

HAD-scale

(baseline diff)

Kääpä 2006

Definitely yes;

Reason: The

randomization list was generated by an independent biostatistician

using a table of random numbers

Definitely yes;

Reason: All

the envelopes were numbered consecutively to avoid rearrangement

of the order.

Physiotherapist opened an opaque sealed envelope

Probably no;

Reason: patient and care providers were (probably) not blinded for group allocation

Probably yes;

Reason: comparable (low) dropout rate in both groups. No imputation for missing questionnaire. values

Definitely yes;

Reason: All relevant outcomes were reported

Definitely yes;

Reason: No other problems noted

Some concerns

sick leave

(afraid to go on sick leave; besides, probably not very sensitive)

Roche (2007; 2011)

Definitely yes;

Reason:

Block randomization

was undertaken with an 8-element permutation

table established by an independent methodologist.

No information;

Probably no;

Reason: physiotherapist who performed outcome assessment was not blinded. No information provided about blinding of patients.

Definitely no;

Reason: more patients in AIT than FRP missing at 1-year follow-up (23% vs. 6%)

Lost to follow-up patients were more frequently on sick leave at baseline, higher number of sick-leave days

Probably yes;

Reason: all relevant outcomes described in the Methods section are included in the Results

Probably yes;

Reason: no other problems noted

Some concerns

Ronzi (2017)

Probably yes;

Reason: The randomization procedure of participants

was determined by a computer randomization feature

that automatically allocated the patient to one of the treatment strategies. This system has been described in previous studies.

Probably yes;

Reason: An independent

research assistant prepared envelopes and numbered

them sequentially according to the randomization

list. Envelopes were given to the

rehabilitation physician who opened them at the end of the clinical evaluations.

Definitely no;

Reason: Participants, therapists, and researchers could not be blinded for the allocated treatment after randomization.

Definitely no;

Reason: 4/49 patients (85) in FRP group lost to follow-up, 23/54 patients (43%) in AIP group lost to follow-up

Probably yes;

Reason: pain, anxiety, depression not mentioned in trial register but were included in study protocol as ‘complementary outcomes’, they were not endpoints of treatment strategies.

Probably no;

Reason: baseline differences disability, pain, number of sick-leave days

HIGH

Tavafian (2011, 2014, 2017)

Probably yes;

Reason: Participants were randomly assigned into the intervention or control group through random permutation blocking of every 6 participants

Probably yes;

Reason: The sequence of allocation was concealed to the rheumatologist who selected the eligible patients. The person who was responsible for random allocation of the eligible selected patients was blind to the clinical and demographic characteristics of the patients.

Probably no;

Reason: The physician and statistical analyst were blinded to the group assignment. The patients were instructed to say nothing about their group assignment to the physician. However, no further measures were undertaken to insure blinding of the physician.

Probably yes;

Reasons: At three months; infrequent loss-to-follow-up (I:5/97; C:4/100)

24 mo: An intention-to-treat analysis using repeated measures analysis of variance (ANOVA) was used to compare the two groups over time. Differences in baseline characteristics of those providing 24-month follow-up and those who did not were assessed to determine any selection bias. Subsequent analyses were restricted to those participants providing 24-month follow-up only. Characteristics of the sample providing this follow-up were compared by treatment allocation. The mean differences of each time point of 3-, 6-, 12- and 24-month follow-up from baseline with 95% confidence intervals (CIs) was computed. Then repeated measures ANOVA was done to examine within- and between-group changes and thus determining the pattern of response over time.

Definitely yes;

Reasons: outcomes according to protocol

Probably yes;

Reason: no other problems noted

Some concerns

Castel (2013)

Definitely yes;

Reason: Patients were randomly assigned in a 1:1

ratio in blocks of 32 according to a computer-generated

random number table

No information;

Probably no;

Reason:

The participants and therapists were not blinded.

Outcome assessors were blinded

Probably no;

Reason: Considerable loss to follow-up and there is a difference between the groups

Post-treatment (I 10% / C 7%)

3 mo (I 10%; C 20%)

6 mo (I 12%; C 24%)

12 mo (I 35%; C 53%)

Probably yes;

Reason: all relevant outcomes described in the Methods section are included in the Results

Probably yes;

Reason: no other problems noted

Some concerns

Van Eijk-Hustings (2013)

Definitely yes;

Reason: computer-generated random numbers

Probably yes;

Reason: opaque, sealed envelopes

Probably no;

Reasons:

There was no blinding to treatment of therapists or participants. Patients in the UC group were not informed about any intervention. Only those who were randomised to MD or AE were invited to participate in the intervention without being informed about the alternative treatment conditions

Definitely no;

Reasons: 67 started MD (despite 108 being randomised to MD), 60 completed MD. Only 19 started AE (despite 47 being randomised to AE), only 8 completed AE. No loss to fu in UC group.

Definitely yes;

Reason: all outcomes described in the Methods section are reported

Despite randomisation, the MD group turned out to be the group with the worst condition at inflow and thus had the largest potential for improvement.

Study conducted both ITT and PP analyses, but loss-to-follow-up may be too large.

HIGH

(HR-QoL , FIQ-total)

Martín (2011, 2014a, 2014b)

Definitely yes;

Reason: a list of random numbers was developed by the statistician

Definitely no;

Reason: open random allocation by a list of random numbers

Probably no;

Reason: data was collected by a researcher not involved in providing treatment.

Unclear whether patients were aware of group allocation

Definitely no;

Reason: relatively high loss to follow-up at 6mo. resp. 40% (I) and 38% (C).

Different dropout rate at assignment/during intervention/follow-up

Definitely yes;

Reason: All relevant outcomes were reported

1) participants were selected from patients referred to a hospital pain management unit (sample might experience greater impact of FM on HRQOL than patients treated in primary care),
2) females were over-represented in our sample.

High risk

FIQ:

Indication of regression to the mean.

Skouen (2006)

Probably yes;

Reason:

The allocation

sequence was prepared beforehand by a physician outside

the clinic. No stratification was

not performed. For each group of 60 patients of the

total of 664, 25 were randomly assigned to treatment as-

usual, 20 to the light program and 15 to the extensive

program. (block randomization due to limited capacity

at the clinic)

No information;

No information;

Probably yes;

Reason: infrequent loss to follow up (I: 5%; C 3%)

Probably yes;

Reason: relevant outcome was reported

Probably no;

This sample of patients with chronic widespread pain was part of a larger study among patients with different kinds of long-term musculoskeletal pain. Was this a pre-planned subgroup analysis?

Some concerns

Becker (2000)

No information;

Probably yes;

Reason: sealed envelopes

Definitely no;

Reason: patients were not blinded and statistical analyses were not performed blind to treatment condition

Probably yes;

Reason: Two patients in the MPT-group were excluded because of manic-depressive disorders, and five patients did not return questionnaires. In the GP-group four patients were excluded for ethical reasons and given MPT, and 12 patients did not return questionnaires. Patients who were excluded or dropped out of the study did not differ from patients included in the study with respect to demographic data, pain epidemiology or HRQL at referral.

Probably yes;

Reason: all outcomes described in the Methods section are reported, no trial registration or study protocol available.

Probably no;

Reason: This study included patients suffering from chronic non-malignant pain conditions, which included a number of participants with other pain than

musculoskeletal pain (e.g. abdominal pain)

Some concerns

Turner-Stokes (2003)

Probably yes;

Reason: a weighted randomization procedure was followed

No information;

Definitely no;

Reason: full blinding not possible. Both therapists and participants were aware of group allocation

Probably no;

Reason:

Dropout rate during treatment was 10 %. Total dropout rate at follow-up assessment was 33%.

Not-at-random dropout of spinal pain patients, but likely the same in both groups

Definitely yes;

Reason: All relevant outcomes were reported

Definitely yes;

Reason: No other problems noted

Some concerns

Pain

(baseline imbalance in pain duration (but corrected for))