Martínez Calderón

Pain Medicine, 22(2), 2021, 481–498
doi: 10.1093/pm/pnaa331
Advance Access Publication Date: 29 September 2020
Review Article
Intervention Therapies to Reduce Pain-Related Fear in Fibromyalgia

Syndrome: A Systematic Review of Randomized Clinical Trials
Javier Martinez-Calderon, PhD*,† Mar Flores-Cortes , PT* Jose Miguel Morales-Asencio, PhD†,‡ and
Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Alejandro Luque-Suarez, PhD*,†
*Department of Physiotherapy, Faculty of Health Sciences, Universidad de Malaga, Malaga, Spain; †Instituto de Investigaci
on Biom
edica de M
alaga
(IBIMA), Malaga, Spain; ‡Department of Nursing, Faculty of Health Sciences, Universidad de Malaga, Malaga, Spain
Correspondence to: Mar Flores-Cortes, PT, Department of Physiotherapy, Faculty of Health Sciences, Universidad de Malaga, Arquitecto
Francisco Penalosa, 3, 29071 Malaga, Spain (Tel: þ 00 34 951 952 853; E-mail: [email protected]), and Alejandro Luque-Suarez, PhD,
Department of Physiotherapy, Faculty of Health Sciences, Universidad de Malaga, Arquitecto Francisco Penalosa, 3, 29071 Malaga, Spain (Tel:
þ 0034 951 952 853; Fax: þ 0034 951 952 853; E-mail: [email protected]).
Disclosure and conflicts of interest: Dr. Javier Martinez-Calderon is supported by the Universidad de Malaga through a postdoctoral grant. All authors state that
they have no conflicts of interest to declare. The funding source was not involved in the conduct of the research or preparation of the article.
Study registration: Open Science Framework (https://osf.io/8s5z3).
Device statement: The article does not contain information about medical device(s).
Data sharing: All data relevant to the study are included in the article or are available as supplementary files.
Patient and public involvement in the research: Patients/athletes/public partners were not involved.
Abstract
Objective. This systematic review aimed to evaluate the effectiveness of different interventions at reducing pain-related fear
in people with fibromyalgia and to analyze whether the included trials reported their interventions in full detail. Design.
Systematic review. Setting. No restrictions. Methods. The Cochrane Library, CINAHL, EMBASE, PsycINFO, PubMed,
and Scopus were searched from their inception to April 2020, along with manual searches and a gray literature
search. Randomized clinical trials were included if they assessed pain-related fear constructs as the primary or sec-
ondary outcome in adults with fibromyalgia. Two reviewers independently performed the study selection, data ex-
traction, risk-of-bias assessment, Template for Intervention Description and Replication (TIDieR) checklist assess-
ment, and grading the quality of evidence. Results. Twelve randomized clinical trials satisfied the eligibility criteria,
including 11 cohorts with a total sample of 1,441 participants. Exercise, multicomponent, and psychological inter-
ventions were more effective than controls were in reducing kinesiophobia. However, there were no differences in
decreasing kinesiophobia when self-management and electrotherapy were used. There were also no differences be-
tween groups with regard to the rest of the interventions and pain-related constructs (fear-avoidance beliefs, fear of
pain, and pain-related anxiety). However, a serious risk of bias and a very serious risk of imprecision were detected
across the included trials. This caused the overall certainty of the judged evidence to be low and very low.
Additionally, the included trials reported insufficient details to allow the full replication of their interventions.
Conclusions. This systematic review shows that there are promising interventions, such as exercise, multicomponent, and
psychological therapies, that may decrease one specific type of fear in people with fibromyalgia, i.e., kinesiophobia. However,
because of the low–very low certainty of the evidence found, a call for action is needed to improve the quality of randomized
clinical trials, which will lead to more definitive information about the clinical efficacy of interventions in this field.
Key words: Chronic Pain; Fibromyalgia; Fear; Review
C The Author(s) 2020. Published by Oxford University Press on behalf of the American Academy of Pain Medicine.
V
All rights reserved. For permissions, please e-mail: [email protected] 481
482 Martinez-Calderon et al.
Introduction online public database (http://doi.org/10.17605/OSF.IO/

Fibromyalgia is associated with a considerable socioeco- 8S5Z3) before data collection and synthesis. JMC and
nomic burden [1–3]. This condition is associated with MFC independently conducted the trial selection, data
structural and functional brain changes [4], central sensi- extraction, risk-of-bias assessment, the Template for
tization symptoms (e.g., allodynia) [5], chronic wide- Intervention Description and Replication (TIDieR)
spread pain [6], physical and mental fatigue [7], sleep checklist assessment, and the evidence grading. All meth-
problems [8], depression [9], suicidal behaviors [10], and odological processes were conducted by consensus. A
death [11]. Fibromyalgia is also linked to other medical third reviewer did not participate in the methodological
and psychiatric disorders, such as irritable bowel syn- process.
drome, migraine, and substance abuse [12–14]. Recent
efforts have been made to establish precise diagnostic cri- Data Sources and Search Strategy
teria for fibromyalgia [6]. However, current interven- The Cochrane Library, CINAHL, EMBASE, PsycINFO,

tions remain unresolved in determining the broad list of PubMed, and Scopus were searched from inception to
symptoms involved in fibromyalgia. This situation fosters April 2020 by using MeSH terms and keywords gener-
feelings of frustration and hopelessness among health ated from subject headings. The full systematic search
professionals and people with fibromyalgia [15]. strategy for each database is listed in the Supplementary
The invisibility of many of the symptoms often leads Data. Gray literature was also looked up via Open Grey
to patients’ feeling stigmatized and misunderstood with and Google Scholar [34]. Pain and rheumatology jour-
regard to their illness process [16]. Patients’ uncertainties nals were also searched, as were the reference lists of the
and fears about the course of fibromyalgia are frequent included trials. Only trials that were written in Spanish
[3, 17, 18]. Generally, fear is a key factor in the manage- or English were considered for inclusion. There were no
ment of chronic musculoskeletal pain and pain-related restrictions in terms of ethnicity, setting, and gender.
symptoms (e.g., pain interference) [19, 20]. Specifically,
a large number of observational studies in fibromyalgia Eligibility Criteria
have found that pain-related fear is associated with The inclusion criteria were:
greater pain intensity [21], functional disability [22], an
1. Studies: randomized clinical trials.
insecure attachment style [23], anxiety [24], and depres-
sion [24]. In the clinical setting, fear of exacerbating fi- 2. Participants: adults with fibromyalgia according to the
bromyalgia symptoms is considered an obstacle to International Association for the Study of Pain (IASP) classifi-
practice exercise modalities, such as yoga [25, 26]. Pain- cation of chronic pain for the International Classification of
related fear has been also shown to modulate the effects Diseases, 11th edition [35].
of pain neuroscience education (PNE) on pain cata-
strophizing [27] and partially mediates changes in muscle 3. Interventions: any experimental intervention, as compared
strength after endurance exercise [28]. with 1) no intervention, 2) waiting list control, 3) sham con-
A number of interventions, such as pain education trol, 4) usual care control, or 5) active control.
[29], cognitive-behavioral therapy [30], and exercise
[31], have decreased pain-related fear in people with fi- 4. Outcome: pain-related fear constructs (fear-avoidance beliefs,
bromyalgia. However, a systematic review evaluating the fear of movement, fear of pain, kinesiophobia, and pain-
level of the evidence for interventions that aim to de- related anxiety).
crease this fear has not yet been conducted. A systematic
review could inform health professionals about the best The exclusion criteria were:
evidence-based interventions currently available in the 1. Trials in which statistical analyses were not performed sepa-
field. Hence, we sought to evaluate which interventions rately for chronic pain conditions.
reduce pain-related fear among individuals with fibromy-
algia by conducting a systematic review of randomized 2. Trials in which pain-related fear values were reported exclu-
clinical trials. Furthermore, we aimed to investigate sively at baseline.
whether the included randomized clinical trials reported
their interventions in full detail. 3. Protocols for randomized clinical trials.
4. Feasibility and pilot trials.

Methods
5. Trials for which the full text was unavailable.
Protocol and Registration
The Cochrane Handbook and the Preferred Reporting
Items for Systematic Reviews and Meta-Analyses Trial Selection
(PRISMA) guidelines were followed [32, 33]. The review First, duplicates were eliminated by using citation man-
protocol was registered in the Open Science Framework agement software (Mendeley desktop v1.19.6) and a
Pain-Related Fear and Fibromyalgia 483
manual check [36], after which the reference list was outcome data, 4) measurement of the outcome, and 5) se-
screened by title and abstract. Second, the full text was lection of the reported results. The responses to the algo-
screened when the trial selection was unclear from read- rithms of the Cochrane risk-of-bias tool (version 2) were
ing titles and abstracts. The trial selection process is listed mapped to judge each risk-of-bias domain and decide the
in Figure 1. overall risk of bias for each included trial [37]. A trial
was determined to have a “low risk of bias” if all bias
Data Extraction domains presented a low risk of bias. A trial was judged
The following information was extracted from each in- to have “some concerns” when one or more bias domains
cluded trial: study year, country, diagnostic criteria for fi- showed some concerns. Finally, a trial was evaluated to
bromyalgia, trial setting, sociodemographic characteristics have a “high risk of bias” if at least one bias domain pre-
(age and gender), clinical characteristics (pain intensity sented a high risk of bias or when multiple domains were
and symptom duration), experimental and control group evaluated to have some concerns [37]. For the present

details, outcome details, and main findings. study, we decided to consider “multiple domains” to
mean that more than 50% of the domains were judged to
have some concerns.
Risk-of-Bias Assessment
Version 2 of the Cochrane risk-of-bias tool was used to
evaluate the risk of bias [37]. This tool has been recom- Intervention Categories
mended to evaluate the risk of bias in randomized clinical Interventions were labeled as 1) “exercise therapy” when
trials [38]. The following domains were examined to de- they were based on exercise modalities designed to ad-
termine whether they had a low, unclear, or high risk of dress pain [39]; 2) “psychological therapy” when they
bias for each included trial: 1) randomization process, 2) were designed to modify behaviors, cognitions, or emo-
deviations from intended interventions, 3) missing tions in response to pain; 3) “self-management therapy”
Identification
1,673 references imported from screening 25 additional references identified through
(557 from Scopus; 403 from EMBASE; 233 other sources (grey literature and manual
from PsycINFO; 211 from PubMed; 135 from searches)
Cochrane Library; 134 from CINAHL)
659 studies screened by title and abstract after

Screening
removing duplicates
585 excluded
62 studies excluded for the

following reasons
1. Not randomised clinical trials

(32)
Eligibility 2. Protocol or feasibility trials (8)

74 studies screened for full-text eligibility 3. Trials where statistical analyses
were not separately performed for
pain conditions (2)
4. No adults with fibromyalgia (1)
5. Pain-related fear only assessed at

baseline (1)
6. The full text was unavailable

(13)
7. Fear analysed inside a broad

Included
psychological construct (e.g.
12 randomized clinical trials were included in the negative mood) (2)
qualitative synthesis, including 11 cohorts
8. Fear constructs not related to
pain (2)
9. Written other language than

English or Spanish (1)
Figure 1. PRISMA flow diagram for study selection.

when they were based only on educational or self- evidence can be graded as high, moderate, low, and very
management strategies aimed to enhance patients’ self- low [56]. Randomized clinical trials start with high qual-
confidence in response to a chronic condition [40]; 4) ity of evidence [57] and can be downgraded one level (se-
“electrotherapy” when they were designed to manage rious) or two levels (very serious) for the presence of risk
pain through the application of different types of cur- of bias, inconsistency, indirectness, imprecision, and pub-
rents; and 5)“multicomponent therapy” when they in- lication bias [57]. Finally, the results were presented in
volved a set of different therapeutic strategies to manage the qualitative synthesis grouped by intervention cate-
pain, e.g., psychological therapy plus exercise. gory and pain-related fear outcome (e.g., psychological
therapy and pain-related anxiety) from the greatest to
lowest certainty of the evidence.
Data Synthesis and Analysis
When any cohort was included in multiple publications,

we performed data extraction, risk-of-bias assessment, Description of Interventions
the TIDieR checklist, and the Grading of The TIDieR checklist was used to analyze whether the in-
Recommendations Assessment, Development and cluded trials reported their experimental interventions in
Evaluation (GRADE) approach exclusively from the old- full detail, in terms of “why” (theoretical framework),
est publication to avoid double-counting participants. “what” (intervention and control description), “who”
Pain-related fear details were extracted from all included (intervention provider), “how” (use of technology, indi-
publications only when pain-related fear was assessed at vidual or group sessions), “where” (setting), “when and
different time points. This step was required only for the how much” (duration, number of sessions), “tailoring,”
publications by McBeth et al. [41] and Beasley et al. [42] and “how well” (fidelity/adherence) [58].
because both involved the same cohort. We planned, on
the basis of our review protocol, to conduct several meta-
analyses to pool the results for each intervention category
and pain-related fear outcome, e.g., exercise and fear- Results
avoidance beliefs. However, we found some reasons The electronic databases retrieved 1,673 citations. A to-
according to the Cochrane Handbook [43]: 1) limited ev- tal of 25 citations were added through manual searches
idence for a pre-specified comparison as a consequence and gray literature. After the removal of duplicates, 659
of the scarce number of trials included in each interven- titles and abstracts were screened. Finally, 74 full texts
tion category, and 2) the impossibility of pooling the were screened after titles and abstracts had been read. A
results when the same outcome and intervention coin- total of 12 randomized clinical trials (1,441 participants,
cided in several included trials because the outcome was 88% female), including 11 cohorts, satisfied the eligibil-
assessed at different follow-up time points [43]. ity criteria. The trial selection process can be followed in
Therefore, the Synthesis Without Meta-analysis Figure 1. A list of the excluded studies in the last screen-
reporting guideline [44] was used to summarize the ing (n ¼ 62) can be found in the Supplementary Data.
results in the present systematic review. The Synthesis The overall sample size ranged from 17 to 442 partici-
Without Meta-analysis guideline promotes transparent pants. The mean age ranged from 39 to 56 years. The
reporting for reviews of interventions through the use of persistence of symptoms ranged from less than 5 years to
a synthesis method as an alternative to meta-analyses 23 years. Electrotherapy, exercise, multicomponent, psy-
[44]. In this context, the included trials were grouped by chological, and self-management interventions were used
intervention category and specific pain-related fear con- as intervention groups. Active interventions, no interven-
structs (e.g., psychological therapy and kinesiophobia) tion, placebo, usual care, and waiting list were used as
from the lowest to greatest risk-of-bias level (see Table control groups. Most of the trials assessed pain-related
1). Furthermore, Cohen’s d with 95% confidence inter- fear as a secondary outcome, and fear-avoidance beliefs,
vals (CIs) and P values were used to present the effect fear of pain, kinesiophobia, and pain-related anxiety as
sizes for each included trial. We used the effect size calcu- primary outcome measures. The Tampa Scale for
lator provided by the Campbell Collaboration to calcu- Kinesiophobia was the most common patient-reported
late effect sizes by using reported means and standard outcome measure, used by six trials, followed by the Pain
deviations for each included trial [52]. P values were Anxiety Symptoms Scale–20, the Fear-Avoidance Beliefs
obtained from 95% CIs following previous recommenda- Questionnaire, and the Fear of Pain Questionnaire,
tions [53]. The magnitude of the effect size was inter- which were used by three trials, two trials, and one trial,
preted by using the benchmark proposed by Cohen [54]: respectively. Some trials conducted their research via tele-
small effect, Cohen’s d ¼ 0.2; medium effect, Cohen’s phone/Internet, whereas others used the following set-
d ¼ 0.5; large effect, Cohen’s d ¼ 0.8 (47). The GRADE tings: laboratory, university facilities, physiotherapy
approach was applied to judge the overall certainty of clinics, and local facilities, such as gymnasiums and fibro-
the evidence for each pain-related fear outcome [55]. In myalgia centers. Further details about each included trial
the GRADE approach, the overall certainty of the are listed in Table 1.
Table 1. Characteristics of included trials
Sample Characteristics Duration/Contact Fear Characteristics
Type of
Study Intervention
Characteristics and Experimental (Experimental Experimental Treatment Follow-Up Experimental
Setting Risk of Bias Group Control Group Group) Group Control Group Facilitator/ Format Assessments Group Control Group Main Findings
Dailey et al. [45] Some concerns of bias Experimental group: [1] Control group: pla- Electrotherapy Active TENS: 60 min [1] Placebo TENS: the Treatment facilitator: (T0) at baseline; (T1) Kinesiophobia (the Kinesiophobia (the Analyses adjusted for
(USA, 2020) n¼103 (103f); age cebo group n¼99 in the clinical set- unit had an identi- TENS allocator post-treatment (af- Tampa Scale for Tampa Scale for site
Kinesiophobia was 44.7 (SD 14.3) yr; (99f); age 47.2 (SD ting between visit 2 cal appearance to Treatment format: 2 ter 4 wk of home Kinesiophobia) Kinesiophobia) No statistically signifi-
considered a sec- symptom duration 12.6) yr; symptom and 3 (30 min each that of the active individual sessions use of active TENS, Mean baseline (SD) Placebo TENS group cant differences be-
ondary outcome median 7 (range 3 duration median 7 session) TENS unit and de- in clinical setting placebo TENS, or 36.5 (7.7) Mean baseline (SD) tween groups were
The sample fulfilled to 12) yr; pain in- (range 2 to 14) yr; þ livered current for plus individual ses- no TENS) Mean change from 37.1 (8.0) observed
the diagnostic crite- tensity at rest (nu- pain intensity at 120 min/day during 4 45 seconds, ramp- sions at home baseline (95% CI) Mean change from Active TENS vs pla-
ria for fibromyalgia merical rating rest (numerical rat- wk at home while ing down to 0 in to post-treatment baseline (95% CI) cebo TENS: mean
based on the scale): 6.2 (SD 1.5) ing scale): 5.9 (SD participants are do- the last 15 seconds -0.7 (-2.0 to 0.6) to post-treatment difference
American College 1.4) ing physical activity þ -0.3 (-1.7 to 1.0) (95% CI)=-0.4 (-
of Rheumatology [2] Control group: no 120 min/day during 4 No TENS group 2.3 to 1.5), P > .99
1990 criteria TENS group n¼99 wk at home while Mean baseline (SD) Active TENS vs no
Setting: University of (99f); age 48.6 (SD participants are do- 37.4 (8.3) TENS: mean differ-
Iowa and 11.8) yr; symptom ing physical activity Mean change from ence (95% CI)=-0.6
Vanderbilt duration median 7 [2] No TENS: a mock baseline (95% CI) (-2.4 to 1.3), P >
University Medical (range 4 to 15) yr; TENS unit was to post-treatment .99
Center pain intensity at used (unit with at- -0.2 (-1.4 to 1.1)
rest (numerical rat- tached electrodes
ing scale): 6.1 (SD that provided no
1.6) electric current
intensity)
Friesen et al. [30] High risk of bias Experimental group: Control group: n¼30 Psychological Internet-delivered cog- Waiting list Treatment facilitator: (T0) at baseline; (T1) Kinesiophobia (the Kinesiophobia (the Statistically significant
(Canada, 2017) n¼30 (28f); age 49 (29f); age 46 (SD intervention nitive behavioral a guide (a doctor- post-treatment Tampa Scale for Tampa Scale for differences between
Kinesiophobia was (SD 10) yr; symp- 13) yr; symptom pain management ate-level clinical Kinesiophobia-17 Kinesiophobia-17 groups in reducing
considered a sec- tom duration 20 duration 13 (SD course: The Pain psychology gradu- items) items) fear of pain were
ondary outcome (SD 9) yr; pain in- 10) yr; pain inten- Course: the course ate student) who Mean baseline (SD) Mean baseline (SD) observed in favor of
The sample fulfilled tensity (the brief sity (the brief pain was conducted for was supervised by a 38.35 (7.39) 40.57 (5.55) the experimental
the diagnostic for fi- pain inventory– inventory–pain se- 8 wk, including 5 registered psycholo- Mean (SD) post- Mean (SD) post- group
bromyalgia based pain severity sub- verity subscale): lessons plus 5- to gist (telephone-de- treatment treatment Post-treatment: effect
on a diagnosis con- scale): 5.45 (SD 6.02 (SD 1.39) 10-min weekly tele- livered support) 33.87 (6.25) 42.73 (4.64) size Cohen’s d
ducted by a 1.10) phone-delivered Treatment format: in- (95% CI)¼1.61
physician support by a guide dividual sessions (1.03 to 2.19), P
Setting: the interven- (a doctorate-level via the Internet range<0.001 to
tion was conducted clinical psychology 0.014
via Internet graduate student)
who was supervised
by a registered
psychologist
McBeth et al. [41] High risk of bias [1] Experimental Control group: n¼109 Psychological [1] Telephone-deliv- Usual care: partici- Treatment facilitators: (T0) at baseline; (T1) Kinesiophobia (the Kinesiophobia (the All analyses adjusted
(UK, 2012) and group: telephone- (76f); age 56.3 (SD intervention ered cognitive be- pants received usual 4 therapists accred- post-treatment (6 Tampa Scale for Tampa Scale for for age, sex, center,
Beasley et al. delivered cognitive 12.5) yr; symptom Exercise intervention havioral therapy: 1 care from their fam- ited by the British months); (T2) 9- Kinesiophobia-17 Kinesiophobia-17 baseline chronic
(2015) behavioral therapy: duration: (–); base- Multicomponent session of 45-60 ily physician Association for month follow-up items) items) pain grade and gen-
Kinesiophobia was n¼112 (80f); age line pain intensity intervention min plus 1 session Behavior and (after randomiza- Mean baseline (SD) Mean baseline (SD) eral health ques-
considered a sec- 56.5 (SD 13.7) yr; (–) per wk during 7 wk Cognitive tion): (T3) 24- Telephone-delivered 36.2 (5.4) tionnaire scores,
ondary outcome symptom duration: of 30-45 min each Psychotherapies month follow-up cognitive behav- Mean (SD) post- and baseline levels
The sample fulfilled (–); baseline pain session, plus 2 (telephone-deliv- after intervention ioral therapy: 36.0 treatment of the outcome of
the diagnostic for intensity (–) booster sessions (1 ered cognitive be- (5.3) 36.0 (6.8) interest
chronic widespread [2] Experimental session at 3 months havioral therapy); Graded exercise: 35.7 Mean (SD) 9-month Statistically significant
pain based on the group: graded exer- and 1 session at 6 fitness instructor (5.4) follow-up (after differences between
definition used in cise: n¼109 (72f); months after ran- (graded exercise) Combination of both randomization) groups in reducing
the American age 55.3 (SD 12.5) domization) þ interventions: 35.5 36.1 (6.5) kinesiophobia were
College of yr; symptom dura- usual care (5.0) Mean (SD) 24-month observed in favor of
Rheumatology cri- tion: (–); baseline [2] Graded exercise: 6 Mean (SD) post- follow-up post- the combination
teria for pain intensity (–) fitness instructor– treatment treatment group (telephone-
fibromyalgia [3] Experimental led monthly Telephone-delivered 36.2 (7.2) delivered cognitive
Setting: graded exer- group: combination appointments, plus cognitive behavioral therapy
cise: leisure- of both exercise for plus graded
(continued)

Type of
Study Intervention
facility– and gym- interventions: improving cardiore- behavioral therapy: exercise) vs usual
based exercise n¼112 (79f); age spiratory fitness 34.2 (6.3) care at 9-month fol-
program 56.7 (SD 13.4) yr; was advised twice Graded exercise: 33.6 low-up (after ran-
symptom duration: per wk (20-60 min (7.0) domization) and
(–); baseline pain per session) and en- Combination of both 24-month follow-
intensity (–) gagement in interventions: 32.8 up post-treatment
“everyday” activi- (6.7) 9-month follow-up (af-
ties during no gym Mean (SD) 9-month ter randomization)
days þ usual care follow-up (after b coefficient
[3] Combination of randomization) (95% CI)=-3.6 (-
both interventions Telephone-delivered 5.2 to -1.9),
as described above cognitive behav- P<0.001
(telephone-deliv- ioral therapy: 34.1 24-month follow-up
ered cognitive be- (6.5) post-treatment
havioral therapy Graded exercise: 33.8 unstandardized b
plus graded (6.7) coefficient
exercise) Combination of both (95% CI)=-2.6 (-
interventions: 32.0 4.3 to -0.9),
(7.2) P<0.004
Mean (SD) 24-month Statistically significant
follow-up post- differences between
treatment groups in reducing
Telephone-delivered kinesiophobia were
cognitive behav- observed in favor of
ioral therapy: 33.3 the telephone-deliv-
(5.7) ered cognitive be-
Graded exercise: 34.1 havioral therapy
(7.6) group vs usual care
Combination of both at 24-month fol-
interventions: 32.9 low-up post-
(7.8) treatment
Unstandardized b coef-
ficient (95% CI)=-
1.9 (-3.6 to -0.2),
P<0.05
No statistically signifi-
cant differences be-
tween groups were
observed for the
rest of comparison
Telephone-delivered
cognitive behav-
ioral therapy vs
usual care
Post-treatment: b coef-
ficient (95% CI)=-
0.9 (-2.3 to 0.9)
9-month follow-up (af-
ter randomization):
b coefficient
(95% CI)=-1.8 (-
3.6 to -0.1)
Graded exercise vs
usual care
ficient (95% CI)=-
1.0 (-2.8 to 0.7)
9-month follow-up af-
ter randomization):
(continued)

Type of
Study Intervention
b coefficient
(95% CI)=-1.9 (-
3.5 to -0.2)
24-month follow-up
post-treatment
Unstandardized b coef-
ficient (95% CI)=-
1.2 (-2.9 to 0.5)
Combination (tele-
phone-delivered
cognitive behav-
ioral therapy plus
graded exercise) vs
usual care
ficient (95% CI)=-
1.8 (-3.5 to -0.1)
Simister et al. [46] High risk of bias Experimental group: Control group: n¼ 34; Psychological Online Acceptance Usual care: partici- Treatment facilitator: (T0) at baseline; (T1) Kinesiophobia (the Kinesiophobia (the Statistically significant
(Canada, 2018) n¼ 33; baseline baseline pain inten- intervention and Commitment pants continued psychologist post-treatment; Tampa Scale for Tampa Scale for differences between
Kinesiophobia was pain intensity (–) sity (–) Therapy (7 modules with their current Treatment format: in- (T2) 3-month fol- Kinesiophobia-11 Kinesiophobia-11 groups in reducing
considered a sec- Total sample: 64f; age with a total of 2 treatment regime dividual sessions low-up items) items) kinesiophobia were
ondary process 39.7 (SD 9.36) yr; months to complete (e.g., medication/ via the Internet Mean baseline (SD) Mean baseline (SD) observed in favor of
The sample fulfilled symptom duration them. Participants spinal nerve blocks, 27.77 (5.97) 27.03 (5.97) the experimental
the 2010 American (since from fibro- were encouraged to massage, physio- Mean (SD) post- Mean (SD) post- group (P<0.001)
College of myalgia diagnosis): dedicate 1 wk to therapy, formal and treatment treatment Post-treatment: effect
Rheumatology clas- 10.16 (SD 7.83) yr complete each mod- informal exercise 21.71 (6.23) 27.45 (5.97) size Cohen’s d
sification criteria ule) þ usual care programs, acupunc- Mean (SD) 3-month Mean (SD) 3-month (95% CI)¼0.95
for fibromyalgia. ture, heat/cold ther- follow-up follow-up (0.41 to 1.47)
Additionally, a apy, and making 23.27 (6.41) 27.40 (6.59) 3-month follow-up: ef-
qualified medical dietary changes) fect size Cohen’s d
professional verified (95% CI)¼0.64
the fibromyalgia (0.12 to 1.15)
diagnosis
Setting: the main inter-
vention was con-
ducted via Internet
van Koulil et al. [47] High risk of bias Experimental group: Control group: pain Multicomponent Cognitive behavioral Waiting list Treatment facilitator: (T0) at baseline; (T1) Kinesiophobia (the Kinesiophobia (the Statistically significant
(The Netherlands, pain avoidance avoidance group: intervention therapy plus exer- psychotherapist post-treatment; Tampa Scale for Tampa Scale for differences between
2011) group: n¼29 (27f); n¼45 (43f); age cise (aerobic, anaer- and social worker (T2) 6-month fol- Kinesiophobia) Kinesiophobia) groups in reducing
The study did not age 45.3 (SD 12.4) 39.4 (SD 10.4) yr; obic, hydrotherapy, (cognitive behav- low-up Mean baseline (SD) Mean baseline (SD) kinesiophobia were
clearly state yr; symptom dura- symptom duration: and relaxation ioral therapy); Pain avoidance group: Pain avoidance group: observed in favor of
whether kinesio- tion: (<5 yr); base- (<5 yr); baseline exercises): 2 ses- physiotherapists 28.5 (6.8) 28.0 (7.0) the experimental
phobia was a pri- line pain intensity pain intensity (6- sions per wk during (exercise) Pain persistence group: Pain persistence group: group at post-treat-
mary or secondary (6-item pain scale): item pain scale): 16 wk (first 120 Treatment format: 23.3 (5.9) 24.0 (5.8) ment and 6-month
outcome 20.3 (SD 2.4) 19.8 (SD 3.1) min of cognitive be- group sessions of Mean (SD) post- Mean (SD) post- follow-up:
The sample fulfilled Experimental group: Control group: pain havioral therapy 8 participants treatment treatment Pain avoidance group:
the 1990 American pain persistence persistence group: followed by 120 Pain avoidance group: Pain avoidance group: F [1,55.89]¼10.59,
College of group: n¼39 (38f); n¼45 (40f); age min of exercise 22.3 (5.3) 26.1 (5.7) P<0.01
Rheumatology clas- age 41.1 (SD 9.4) 40.9 (SD 10.4) yr; each session) plus Pain persistence group: Pain persistence group: 3.34 points lower for
sification criteria yr; symptom dura- symptom duration: 90 min/day of 20.7 (5.0) 23.5 (6.0) kinesiophobia in
for fibromyalgia tion: (<5 yr); base- (<5 yr); baseline homework to sup- Mean (SD) 6-month Mean (SD) 6-month the experimental
Setting: outpatient line pain intensity pain intensity (6- port both interven- follow-up follow-up group compared
treatment program (6-item pain scale): item pain scale): tions. Additionally, Pain avoidance group: Pain avoidance group: with the control
19.1 (SD 3.7) 17.6 (SD 3.4) 1 booster session at 24.3 (6.5) 26.1 (7.4) group
3 months after the Pain persistence group: Pain persistence group: Pain persistence group:
intervention 20.5 (5.7) 24.0 (5.2) F [1,75.53]¼14.31,
P<0.001
2.57 points lower for
kinesiophobia in
(continued)

Type of
Study Intervention
the experimental
group compared
with the control
group
van Oosterwijck et al. High risk of bias Experimental group: Control group: n¼15 Self-management PNE: 2 sessions. 30 Pacing self-manage- Treatment facilitator: (T0) at baseline; (T1) 2 Kinesiophobia (the Kinesiophobia (the No statistically signifi-
[48] (Belgium, n¼15 (12f); age (14f); age 45.9 (SD intervention min in the first ses- ment education: 2 a bachelor in phys- wk follow-up; (T2) Tampa Scale for Tampa Scale for cant differences be-
2013) 45.8 (SD 9.5) yr; 11.5) yr; symptom sion. The second sessions. 30 min in iotherapy who re- 3-month follow-up Kinesiophobia-17 Kinesiophobia-17 tween groups were
Kinesiophobia was symptom duration: duration: 116 (SD session was by tele- the first session. ceived training items) items) observed
considered a sec- 156 (SD 96) 46) months; base- phone, and the du- The second session from 2 qualified Mean baseline (SD) Mean baseline (SD) Effect size¼0.21
ondary outcome months; baseline line pain intensity ration was was by telephone, physiotherapists 38.7 (10.7) 40.7 (8.4)
The sample fulfilled pain intensity (–) (–) unknown. and the duration with experience in Mean (SD) 2-wk fol- Mean (SD) 2-wk fol-
the 1990 American was unknown. providing low-up low-up
College of education 34.9 (10.1) 39.8 (7.1)
Rheumatology clas- Treatment format: in- Mean (SD) 3-month Mean (SD) 3-month
sification criteria dividual sessions follow-up follow-up
for fibromyalgia (one-on-one) 36.5 (9.9) 39.9 (8.2)
Setting: University fa-
cilities in Vrije
Universiteit Brussel
and University
College Antwerp
Amer-Cuenca et al. High risk of bias [1] PNE high-dose Control group: n¼17 Self-management [1] PNE high-dose Biomedical education: Treatment facilitator: (T0) at baseline; (T1) Pain-related anxiety Pain-related anxiety No statistically signifi-
[29] (Spain, 2020) group: n¼20 (17f); (16f); age 51.27 intervention group: 6 sessions of 90 min divided into physiotherapists post-treatment; (Spanish version of (Spanish version of cant differences be-
Pain-related anxiety age 54.75 (SD (SD 10.57) yr; 45 min each session 2 sessions with expertise in (T2) 3-month fol- the Pain Anxiety the Pain Anxiety tween groups and
was considered a 10.14) yr; symptom symptom duration [2] PNE concentrated applying PNE low-up post- Symptoms Scale- Symptoms Scale- time were observed
secondary outcome duration 23.53 (SD 20.92 (SD 9.02) yr; low-dose group: 2 Treatment format: treatment 20) 20) The pain anxiety
The sample fulfilled 13.57) yr; pain in- pain intensity (vi- sessions of 45 min group sessions with Mean baseline (SD) Mean baseline (SD) symptoms subscale
the 1990 and 2010 tensity (visual ana- sual analog scale) each session 4 to 6 participants The Pain Anxiety The Pain Anxiety 1: F¼0.626,
American College log scale) 6.55 (SD 8.30 (SD 1.45) [3] PNE diluted low- Symptoms Scale- Symptoms Scale- P¼0.709,
of Rheumatology 2.27) dose group: 6 ses- 20-subscale 1 20-subscale 1 n2p¼0.029
classification crite- [2] PNE concentrated sions of 15 min PNE high-dose group: 35.73 (15.13) The pain anxiety
ria for fibromyalgia low-dose group: each session 38.41 (14.12) The Pain Anxiety symptoms subscale
Setting: 3 fibromyalgia n¼20 (20f); age PNE concentrated Symptoms Scale- 2: F¼0.882,
centers 55.20 (SD 8.19) yr; low-dose group: 20-subscale 2 P¼0.510,
symptom duration 37.76 (11.91) 13.86 (7.52) n2p¼0.041
23.18 (SD 15.50) PNE diluted low-dose Mean (SD) post-
yr; pain intensity group: 38.94 treatment
(visual analog scale) (12.66) The Pain Anxiety
7.07 (SD 1.43) The Pain Anxiety Symptoms Scale-
[3] PNE diluted low- Symptoms Scale- 20-subscale 1 post-
dose group: n¼20 20-subscale 2 treatment: 32.20
(18f); age 51.67 PNE high-dose group: (12.32)
(SD 7.38) yr; symp- 16.70 (5.39) The Pain Anxiety
tom duration 12.64 PNE concentrated Symptoms Scale-
(SD 10.88) yr; pain low-dose group: 20-subscale 2 post-
intensity (visual an- 14.47 (5.60) treatment: 12.26
alog scale) 7.44 (SD PNE diluted low-dose (6.64)
1.65) group: 14.70 (6.11) Mean (SD) 3-month
Mean (SD) post- follow-up
treatment The Pain Anxiety
The Pain Anxiety Symptoms Scale-
Symptoms Scale- 20-subscale 1 3-
20-subscale 1 post- month follow-up:
treatment 28.53 (15.26)
PNE high-dose group: The Pain Anxiety
35.88 (15.96) Symptoms Scale-
PNE concentrated 20-subscale 2 3-
low-dose group: month follow-up:
35.23 (8.81) 11.53 (8.12)
(continued)

Type of
Study Intervention
PNE diluted low-dose
group: 36.47
(13.16)
The Pain Anxiety
Symptoms Scale-
20-subscale 2 post-
treatment
PNE high-dose group:
14.23 (8.37)
PNE concentrated
low-dose group:
12.88 (4.80)
group: 15.11 (6.68)
Mean (SD) 3-month
follow-up
The Pain Anxiety
Symptoms Scale-
20-subscale 1 3-
month follow-up
33.35 (15.22)
PNE concentrated
low-dose group:
35.82 (13.01)
group: 36.64
(16.13)
The Pain Anxiety
Symptoms Scale-
20-subscale 2 3-
month follow-up
13.88 (6.67)
PNE concentrated
low-dose group:
13.17 (4.86)
group: 15.35 (7.39)
Carleton et al. [49] High risk of bias Total sample: n¼17 (–) Psychological The attentional modifi- Attention control con- (–) (T0) at baseline; (T1) Pain-related anxiety Pain-related anxiety Differences between
(Canada, 2011) (16f); age 51.2 (SD intervention cation paradigm: 2 dition (placebo): 2 post-treatment (the Pain Anxiety (the Pain Anxiety groups were not
Pain-related anxiety 6.0) yr; symptom sessions per wk of sessions per wk of Symptoms Scale- Symptoms Scale- reported
and fear of pain duration 12.3 (SD 15 min each session 15 min each session 20) 20)
were considered 6.2) yr (9 partici- for 4 wk for 4 wk Mean baseline (SD) Mean baseline (SD)
secondary pants in the experi- 27.11 (16.89) 44.50 (22.74)
outcomes mental group and Mean (SD) post- Mean (SD) post-
The sample fulfilled 8 participants in the treatment treatment
the diagnostic crite- control group) 24.44 (15.68) 28.00 (29.87)
ria for fibromyalgia Experimental group: Fear of pain (the fear Fear of pain (the fear
based on a diagnos- pain intensity (vi- of pain question- of pain question-
tic interview from sual analog scale) naire short form) naire short form)
the American experimental Mean baseline (SD) Mean baseline (SD)
College of group: 6.44 (SD 42.00 (14.70) 38.83 (11.58)
Rheumatology and 1.67) Mean (SD) post- Mean (SD) post-
diagnosis from a li- Control group: pain treatment treatment
censed physician intensity (visual an- 36.11 (9.06) 39.33 (13.92)
Setting: laboratory alog scale) 6.00 (SD
2.00)
High risk of bias (–)
(continued)

Type of
Study Intervention
Carleton et al. [50] Experimental group: Control group: n¼54 Psychological The attention bias Standard attention (T0) at baseline; (T1) Pain-related anxiety Pain-related anxiety No statistically signifi-
(Canada, 2020) n¼63 (63f). Pain in- (54f). Pain intensity intervention modification deliv- task: 10 sessions post-treatment; (the Pain Anxiety (the Pain Anxiety cant differences be-
The study did not tensity (visual ana- (visual analog ered through the during a period of 6 (T2) 1-month fol- Symptoms Scale- Symptoms Scale- tween groups were
clearly state if pain- log scale): 58.57 scale): 62.52 (SD Internet (attention wk divided into 2 low-up 20) 20) observed
related anxiety was (SD 18.99). This 18.53). This study task): 10 sessions sessions of 45 min Mean baseline (SD) Mean baseline (SD) Post-treatment: fixed
either a primary or study reported age only reported the during a period of 6 each session and 46.19 (26.54) 49.89 (22.80) effects b
a secondary only for those par- age for those partic- wk divided into 2 8 sessions of 15 min Mean (SD) post- Mean (SD) post- (95% CI)¼2.15 (-
outcome ticipants who com- ipants that com- sessions of 45 min each session treatment treatment 4.87 to 9.94),
The sample fulfilled pleted the study pleted the study each session and 45.86 (25.44) 47.48 (25.39) P¼0.488
the diagnostic crite- (n¼26): 47.92 (SD (n¼29): 47.98 (SD 8 sessions of 15 min Mean (SD) 1-month Mean (SD) 1-month 1-month follow-up:
ria for fibromyalgia 10.75) yr 11.89) yr each session follow-up follow-up fixed effects b
based on the 33.80 (23.95) 49.69 (29.49) (95% CI)=-3.88 (-
American College 14.97 to 7.44),
of Rheumatology P¼0.453
preliminary diag-
nostic criteria for
fibromyalgia
Setting: the interven-
tion was conducted
via Internet
Larsson et al. [51] High risk of bias Experimental group: Control group: n¼63 Exercise intervention Progressive resistance Relaxation therapy: 2 Treatment facilitator: (T0) at baseline; (T1) Fear-avoidance beliefs Fear-avoidance beliefs No statistically signifi-
(Sweden, 2015) n¼67 (67f); age (63f); age 52.10 exercise: 2 sessions sessions per wk supervised by post-treatment; (the Fear- (the Fear- cant differences be-
Fear-avoidance beliefs 50.81 (SD 9.05) yr; (SD 9.78) yr; symp- per wk during 15 during 15 wk physiotherapists (T2) 13- to 18- Avoidance Beliefs Avoidance Beliefs tween groups were
were considered a symptom duration tom duration 9.44 wk Treatment format: month follow-up Questionnaire) Questionnaire) observed
secondary outcome 11.06 (SD 8.53) yr; (SD 7.33) yr; base- group sessions Mean baseline (SD) Mean baseline (SD) Post-treatment fear-
The sample fulfilled baseline pain inten- line pain intensity ranging from 5 to 7 Fear-avoidance beliefs Fear-avoidance beliefs avoidance beliefs
the diagnostic for fi- sity (visual analog (visual analog participants about physical ac- about physical ac- about physical ac-
bromyalgia based scale): 49.3 (SD scale): 52.4 (SD tivity: 9.7 (6.1) tivity: 11.2 (6.1) tivity: P¼0.92
on the American 23.9) 18.3) Fear-avoidance beliefs Fear-avoidance beliefs Post-treatment fear-
College of about work: 17.2 about work: 15.9 avoidance beliefs
Rheumatology (12.7) (12.1) about work:
1990 classification Mean (SD) post- Mean (SD) post- P¼0.79
criteria for treatment treatment 13- to 18-month fol-
fibromyalgia Fear-avoidance beliefs Fear-avoidance beliefs low-up fear-avoid-
Setting: physiotherapy about physical ac- about physical ac- ance beliefs about
premises and a local tivity: 8.9 (6.1) tivity: 10.3 (6.3) physical activity:
gym at 4 different Fear-avoidance beliefs Fear-avoidance beliefs P¼0.38
sites about work: 17.8 about work: 16.67 13- to 18-month fol-
(13.1) (12.5) low-up fear-avoid-
Mean (SD) 13-18- Mean (SD) 13-18- ance beliefs about
month follow-up month follow-up work: P¼0.59
Fear-avoidance beliefs Fear-avoidance beliefs
about physical ac- about physical ac-
tivity: 10.1 (5.2) tivity: 9.7 (6.3)
about work: 15.5 about work: 14.7
(11.5) (12.4)
de Medeiros et al. [31] High risk of bias Experimental group: Control group: n¼21 Exercise intervention Mat Pilates: 2 sessions Aquatic aerobic exer- Treatment facilitator: (T0) at baseline; (T1) Fear-avoidance beliefs Fear-avoidance beliefs No statistically signifi-
(Brazil, 2020) n¼21 (21f); age (21f); age 50.7 (SD per wk during 12 cise: 2 sessions per physiotherapist post-treatment (12 (the Fear-avoidance (the Fear- cant differences be-
Fear-avoidance beliefs 45.5 (SD 10.6) yr; 9.7) yr; symptom wk (50 min each wk during 12 wk Treatment format: wk after Beliefs Avoidance Beliefs tween groups were
were considered a symptom duration: duration: (–); base- session) (40 min each group sessions up randomization) Questionnaire- Questionnaire- observed
secondary outcome (–); baseline pain line pain intensity session) to 4 participants Brazilian version) Brazilian version) Post-treatment (12 wk
The sample fulfilled intensity (visual an- (visual analog Mean baseline (SD) Mean baseline (SD) after randomiza-
the diagnostic for fi- alog scale): 7.5 (SD scale): 7.5 (SD 1.8) Fear-avoidance beliefs Fear-avoidance beliefs tion) fear-avoidance
bromyalgia based 1.6) about physical ac- about physical ac- beliefs about physi-
on the 2010 tivity: 13.0 (7.1) tivity: 12.6 (7.0) cal activity: ad-
American College Fear-avoidance beliefs Fear-avoidance beliefs justed with mean
of Rheumatology about work: 25.0 about work: 27.5 difference
(10.8) (12.6)
(continued)

tion) fear-avoidance
(95% CI)¼2.8 (-2.0
(95% CI)¼4.4 (-3.4

adjusted with mean
beliefs about work:
Post-treatment (12 wk
to 12.3), P¼0.26
after randomiza-
to 7.6), P¼0.24
Main Findings
Risk-of-Bias Assessment
difference
One trial was judged to have bias concerns [45]. The rest
of the included trials were judged to have a high risk of
bias [29–31,41,46–51] (Table 2). Overall, biases in the

Mean (SD) post-treat- Mean (SD) post-treat-
about physical ac-

ment (12 wk after
about work: 30.3

tivity: 10.9 (8.4)
randomization)
randomization process and missing outcome data were
Control Group
the bias domains judged to have less risk of bias. The in-
(11.2)
cluded trials were frequently judged to have biases asso-
ciated with deviations from the intended interventions,
measurements of the outcome, and the selection of the
Fear Characteristics
about physical ac-

ment (12 wk after
about work: 25.8

reported results (Table 2 and Figure 2).
randomization)
tivity: 8.1 (6.9)

Experimental
(14.0)
Completeness of Intervention Descriptions

Group

The included trials provided insufficient descriptions of
their experimental interventions, according to the
TIDieR checklist. Trials often reported the name and the
Facilitator/ Format Assessments
rationale of the intervention, doses, and materials and

Follow-Up
procedures applied in each intervention. On the other

hand, items such as 1) information related to tailoring
the intervention, 2) fidelity/adherence, and 3) the possi-
ble modifications of the intervention were not provided
in detail (Figure 3).
Treatment
The Overall Certainty of the Evidence: The GRADE

Approach
Control Group
The overall certainty of the available evidence on the ef-

fectiveness of psychological therapy and exercise inter-
vention in reducing kinesiophobia was low because of
the presence of a very serious risk of bias. The certainty
of the available evidence on the effectiveness of electro-
Duration/Contact
therapy in decreasing kinesiophobia was low because of

Experimental
the presence of a serious risk of bias and imprecision.

Finally, the certainty of the available evidence on the ef-
Group
fectiveness of the rest of interventions in reducing kinesi-

ophobia and other pain-related fear outcomes, such as
pain-related anxiety, fear-avoidance beliefs, and fear of
(Experimental
Intervention
pain, was judged to be very low because of a very serious

Type of
Group)
risk of bias and imprecision. Further details are listed in

Table 3.
wk¼weeks; min¼minutes; yr¼years; f¼female; and SD¼standard deviation.
Control Group
Psychological Therapy and Kinesiophobia (Low

Evidence)
Internet-delivered acceptance and commitment therapy
Sample Characteristics
was more effective than usual care in reducing kinesio-

phobia at post-treatment (Cohen’s d¼0.94, 95% CI 0.41
to 1.47, P<0.001; large effect) and 3-month follow-up
Experimental
(Cohen’s d¼0.63, 95% CI 0.12 to 1.15, P¼0.01; me-

Group
dium effect) [46]. Furthermore, Internet-delivered cogni-

tive behavioral therapy was more effective than waiting
list in decreasing kinesiophobia at post-treatment
(Cohen’s d¼1.61, 95% CI 1.03 to 2.19, P<0.0001; large
Risk of Bias
effect) [30]. Finally, telephone-delivered cognitive behav-

ioral therapy was more effective than usual care in less-
ening kinesiophobia at 9-month follow-up (after
Characteristics and
randomization) (Cohen’s d¼0.31, 95% CI 0.02 to 0.59,

Setting: a group room
P¼0.03; small effect) [41] and 24-month follow-up

Physiotherapy
classification
of the Clinic
(Cohen’s d¼0.44, 95% CI 0.14 to 0.74, P¼0.004; small

criteria
School
Setting
Study
Table 2. Risk-of-bias assessment
Bias in Bias in the

Bias in the Deviation from Bias in Selection of the
Randomization the Intended Bias in Missing Measurement of Reported The Overall
Author(s) and Year Process Intervention Outcome Data the Outcome Results Risk of Bias
Amer-Cuenca et al 2020 [29] L H H L SC H
Carleton et al. 2011 [49] SC SC H L SC H
Carleton et al. 2020 [50] SC H H L SC H
Dailey et al. 2020 [45] L SC L L L SC
Friesen et al. 2017 [30] L SC L H SC H
Larsson et al. 2015 [51] L SC SC H SC H
McBeth et al. 2012 [41] SC SC L H SC H
de Medeiros et al. 2020 [31] L SC L H SC H
Simister et al. 2018 [46] L SC L H SC H
van Koulil et al. 2011 [47] H SC SC H SC H

van Oosterwijck et al. 2013 [48] SC SC L L SC H
H¼high risk of bias; SC¼some concerns of bias; L¼low risk of bias.
effect) [42]. However, there were no differences between Electrotherapy and Kinesiophobia (Low Evidence)
groups at post-treatment (Cohen’s d¼0.27, 95% CI - Active transcutaneous electrical nerve stimulation
0.02 to 0.57, P¼0.07) [41]. The inconsistency found be- (TENS) was not more effective than either placebo TENS
tween the result obtained by McBeth et al. (b coeffi- or no TENS in reducing kinesiophobia at post-treatment
cient=-1.8, 95% CI -3.6 to -0.1) [41] and the result (active TENS vs placebo TENS, Cohen’s d¼0.09,
obtained when Cohen’s d was calculated (Cohen’s 95% CI -0.19 to 0.37, P¼0.54; active TENS vs no
d¼0.31, 95% CI 0.02 to 0.59, P¼0.03) at 9-month TENS, Cohen’s d¼0.21, 95% CI -0.06 to 0.49, P¼0.13)
follow-up (after randomization) could be clarified by see- [45].
ing that McBeth et al. applied a multivariate analysis
adjusting for several factors, such as age, sex, center, Self-Management Intervention and Kinesiophobia
baseline chronic pain grade and general health question- (Very Low Evidence)
naire scores, and baseline levels of the outcome of inter- There were not statistically significant differences be-
est. That may explain why they found no statistically tween PNE and pacing self-management education with
significant differences between groups. regard to effectiveness at decreasing kinesiophobia at 2-
week follow-up (Cohen’s d¼0.56, 95% CI -0.17 to 1.29,
Exercise Intervention and Kinesiophobia (Low P¼0.13) and 3-month follow-up (Cohen’s d¼0.37,
Evidence) 95% CI -0.35 to 1.1, P¼0.32) [48].
Graded exercise was more effective than usual care in
diminishing kinesiophobia at post-treatment (Cohen’s Multicomponent Intervention and Kinesiophobia
d¼0.35, 95% CI 0.05 to 0.64, P¼0.02; small effect) and (Very Low Evidence)
9-month follow-up (after randomization) (Cohen’s Telephone-delivered cognitive behavioral therapy plus
d¼0.35, 95% CI 0.07 to 0.63, P¼0.01; small effect) graded exercise was more effective than usual care in
[41]. However, these effects were not maintained at 24- lessening kinesiophobia at post-treatment (Cohen’s
month follow-up (Cohen’s d¼0.28, 95% CI -0.01 to d¼0.47, 95% CI 0.18 to 0.77, P¼0.002; small effect)
0.57, P¼0.057) [42]. The inconsistency found between [41], 9-month follow-up (after randomization) (Cohen’s
the result obtained by McBeth et al. at post-treatment (b d¼0.60, 95% CI 0.31 to 0.88, P < 0.0001; medium ef-
coefficient=-1.0, 95% CI -2.8 to 0.7) and 9-month fect) [41], and 24-month follow-up (Cohen’s d¼0.44,
follow-up (after randomization) (b coefficient=-1.9, 95% CI 0.15 to 0.73, P¼0.003; small effect) [42].
95% CI -3.5 to -0.2) [41] and the result obtained when Moreover, cognitive behavioral therapy plus a broad set
Cohen’s d was calculated at post-treatment (Cohen’s of exercises (e.g., aerobic or relaxation exercises) was
d¼0.35, 95% CI 0.05 to 0.64, P ¼ 0.02) and 9-month more effective than waiting list in diminishing kinesio-
follow-up (after randomization) (Cohen’s d¼0.35, phobia at post-treatment (pain avoidance group, Cohen’s
95% CI 0.07 to 0.63, P¼0.01) could be explained by see- d¼0.68, 95% CI 0.16 to 1.21, P¼0.01; medium effect;
ing that McBeth et al. applied a multivariate analysis ad- pain persistence group, Cohen’s d¼0.50, 95% CI 0.05 to
justed for several factors, such as age, sex, center, 0.96, P¼0.03; medium effect) [47]. However, the combi-
baseline chronic pain grade and general health question- nation of cognitive behavioral therapy and different types
naire scores, and baseline levels of the outcome of inter- of exercise was more effective than waiting list in reduc-
est. That may explain why they found no statistically ing kinesiophobia only in the pain persistence group at 6-
significant differences between groups. month follow-up (pain avoidance group, Cohen’s
Risk of bias
0% 20% 40% 60% 80% 100%
Bias in randomization process
Bias in deviations from the intended

interventions
Low
Some concerns
Bias in missing outcome data
High
Bias in measurement of the outcome

Bias in selection of the reported results
Figure 2. Risk-of-bias graph: review authors’ judgments about each risk-of-bias domain, presented as percentages across all in-
cluded trials.
11
10
5
Exercise therapy
4
Multicomponent
3 Electrotherapy
2 Psychological therapy
Self-management
1
Figure 3. The TIDieR checklist.
d¼0.25, 95% CI -0.27 to 0.78, P¼0.36; pain persistence baseline levels of the outcome of interest. That may ex-
group, Cohen’s d¼0.64, 95% CI 0.18 to 1.11, plain why they found no statistically significant differ-
P¼0.007; medium effect) [47]. The inconsistency found ences between groups. Furthermore, the inconsistency
between the result obtained by McBeth et al. (b coeffi- found between the result obtained by van Koulil et al.
cient=-1.8, 95% CI -3.5 to -0.1) [41] and the result (F [1,55.89]¼10.59, P<0.01) [47] and the result
obtained when Cohen’s d was calculated (Cohen’s obtained when Cohen’s d was calculated (Cohen’s
d¼0.47, 95% CI 0.18 to 0.77, P¼0.002) at post- d¼0.25, 95% CI -0.27 to 0.78, P¼0.36) at 6-month
treatment can be clarified by seeing that McBeth et al. follow-up can be explained by noting that van Koulil
applied a multivariate analysis adjusting for several fac- et al. calculated a global result for all follow-up time
tors, such as age, sex, center, baseline chronic pain points, whereas Cohen’s d was specifically calculated
grade and general health questionnaire scores, and for each follow-up time point.
Table 3 The overall certainty of the evidence (GRADE)
Summary of Findings The Certainty of the Evidence (Based on the GRADE Approach)
Number
of Participants Risk of Level of
Outcome Studies (N) Bias Inconsistency Indirectness Imprecision Evidence Importance
Kinesiophobia
Psychological therapy 3 569 Very serious* No No No Low Critical
(-2)
† ‡
Electrotherapy 1 301 Serious No No Serious (-1) Low Critical
(-1)
*
Exercise 1 442 Very serious No No No Low Critical
(-2)
* §
Multicomponent 2 600 Very serious No No Serious (-1) Very low Critical
intervention (-2)
* ‡ ,§
Self-management 1 30 Very serious No No Very serious (-2) Very low Critical

(-2)
Pain-related anxiety
Psychological therapy 2 134 Very serious* No No Very serious‡ ,§ Very low Critical
(-2) (-2)
* ‡ ,§
Self-management 1 77 Very serious No No Very serious Very low Critical
(-2) (-2)
Fear-avoidance beliefs
Exercise 2 172 Very serious* No No Very serious‡,§ Very low Critical
(-2) (-2)
Fear of pain
Psychological therapy 1 17 Very serious* No No Very serious‡,§ Very low Critical
(-2) (-2)
GRADE¼Grading of Recommendations Assessment, Development, and Evaluation. Publication bias was not considered because of the scarce number of in-
cluded studies.
*We rated down two levels because all information came from randomized trials with a high risk of bias with potential limitations that are likely to lower confi-
dence in the estimate of effect.
†
We rated down one level because all information came from randomized trials with some concerns of bias with potential limitations that are likely to lower
confidence in the estimate of effect.
‡
We rated down one level because of the presence of serious imprecision associated with a total sample size of less than 400.
§
We rated down one level because of the presence of serious imprecision associated with trials that did not report confidence intervals.
Exercise Intervention and Fear-Avoidance Beliefs Psychological Therapy and Pain-Related Anxiety
(Very Low Evidence) (Very Low Evidence)
There were not differences between mat Pilates and An Internet-delivered attention task was more effective
aquatic exercises with regard to effectiveness at decreas- than a standard attention task in decreasing pain-related
ing fear-avoidance beliefs at post-treatment (physical ac- anxiety at 1-month follow-up (Cohen’s d¼0.59, 95% CI
tivity beliefs, Cohen’s d¼0.36, 95% CI -0.24 to 0.97, 0.04 to 1.13, P¼0.03; medium effect), although no differ-
P¼0.25; work beliefs, Cohen’s d¼0.35, 95% CI -0.25 to ences were observed between groups at post-treatment
0.96, P¼0.26) [31]. Furthermore, there were no differen- (Cohen’s d¼0.06, 95% CI -0.30 to 0.43, P¼0.76) [50].
ces between progressive resistance exercise and relaxa- Moreover, no differences between the attentional modifica-
tion therapy with regard to effectiveness at lessening tion paradigm and the attention control condition were
fear-avoidance beliefs at post-treatment (physical activity found with regard to effectiveness at lessening pain-related
beliefs, Cohen’s d¼0.23, 95% CI -0.16 to 0.61, P¼0.24; anxiety at post-treatment (Cohen’s d¼0.15, 95% CI -0.80
work beliefs, Cohen’s d¼0.09, 95% CI -0.29 to 0.47, to 1.11, P¼0.77) [49]. The inconsistency between the result
P¼0.66) and 13- to 18-month follow-up (physical activ- obtained by Carleton et al. (b=-3.88, 95% CI -14.97 to
ity beliefs, Cohen’s d¼0.07, 95% CI -0.34 to 0.48, 7.44, P¼0.453) [50] and the result attained when Cohen’s
P¼0.75; work beliefs, Cohen’s d¼0.07, 95% CI -0.34 to d was calculated (Cohen’s d¼0.59, 95% CI 0.04 to 1.13,
0.48, P¼0.75) [51]. P¼0.03) at 1-month follow-up can be explained by noting
that Carleton et al. applied a multivariate analysis including
several factors, such as participants and time points. That
may explain why they found no statistically significant dif-
Psychological Therapy and Fear of Pain (Very Low
ferences between groups.
Evidence)
There were no differences between the attentional modi-
fication paradigm and an attention control condition Self-Management Intervention and Pain-Related
with regard to effectiveness at reducing fear of pain at Anxiety (Very Low Evidence)
post-treatment (Cohen’s d¼0.28, 95% CI -0.68 to 1.23, There were no differences between PNE and biomedical
P¼0.58) [49]. education with regard to effectiveness at diminishing
pain-related anxiety at post-treatment (high-dose PNE revealed that different interventions reduce a specific
group Pain Anxiety Symptoms Scale [PASS]-1 subscale, pain-related fear construct (kinesiophobia), the overall
Cohen’s d¼0.25, 95% CI -0.39 to 0.90, P¼0.46; high- certainty of the evidence for all included trials was judged
dose PNE PASS-2 subscale, Cohen’s d¼0.26, 95% CI - to be low and very low after application of the GRADE
0.39 to 0.91, P¼0.44; concentrated low-dose PNE PASS- approach. This poor certainty of the evidence is associ-
1 subscale, Cohen’s d¼0.29, 95% CI -0.36 to 0.94, ated with the presence of a serious risk of bias and impre-
P¼0.39; concentrated low-dose PNE PASS-2 subscale, cision across the trials. Furthermore, the included trials
Cohen’s d¼0.11, 95% CI -0.54 to 0.75, P¼0.75; diluted reported incomplete information about their experimen-
low-dose PNE PASS-1 subscale, Cohen’s d¼0.33, tal interventions. In this context, the serious methodolog-
95% CI -0.32 to 0.98, P¼0.33; diluted low-dose PNE ical flaws found across the included trials preclude
PASS-2 subscale, Cohen’s d¼0.43, 95% CI -0.23 to 1.08, establishing definitive clinical implications.
P¼0.20) and 3-month follow-up (high-dose PNE PASS-1

subscale, Cohen’s d¼0.32, 95% CI -0.33 to 0.97,
Methodological Considerations
P¼0.34; high-dose PNE PASS-2 subscale, Cohen’s
After application of the TIDieR checklist, the included
d¼0.32, 95% CI -0.33 to 0.97, P¼0.34; concentrated
trials were evaluated as delivering insufficient detail
low-dose PNE PASS-1 subscale, Cohen’s d¼0.52,
about their interventions. This issue can make the trans-
95% CI -0.14 to 1.17, P¼0.12; concentrated low-dose
lation of research results to clinical practice difficult [62].
PNE PASS-2 subscale, Cohen’s d¼0.25, 95% CI -0.40 to
For instance, details associated with how well fidelity
0.90, P¼0.46; diluted low-dose PNE PASS-1 subscale,
and adherence were assessed were often insufficient.
Cohen’s d¼0.51, 95% CI -0.14 to 1.17, P¼0.13; diluted
Fidelity/adherence is a key aspect when trials are con-
low-dose PNE PASS-2 subscale, Cohen’s d¼0.49,
ducted. When both factors are evaluated to be poor, in-
95% CI -0.16 to 1.15, P¼0.14) [29].
formation related to this issue facilitates the
interpretation of the results and demonstrates the neces-
Discussion sity of performing further research in the field [63, 64].
Information on how interventions were tailored (or not)
Overall, exercise, multicomponent, and psychological
was also scarce. Multiple factors are involved in the per-
interventions were found to decrease kinesiophobia in
petuation of fibromyalgia symptoms [65]. Therefore,
people with fibromyalgia at different follow-up points, as
interventions should not be identical, and researchers
compared with controls. A previous systematic review
should consider a biopsychosocial perspective when eval-
analyzed which interventions may reduce fear in people
uating their effectiveness [65], or at least they should re-
with chronic low back pain [59]. Interestingly, our pre-
port details about whether their interventions were
sent findings agree with that systematic review, because
tailored or not. The GRADE approach judged the overall
both found that exercise, multicomponent, and psycho-
certainty of the evidence as low and very low across
logical interventions are effective approaches in decreas-
interventions. A serious risk of bias and a very serious
ing kinesiophobia. Both systematic reviews have also
risk of imprecision were detected across the included tri-
revealed that electrotherapy is not an adequate interven-
als. Biases associated with deviations from the intended
tion to decrease kinesiophobia in individuals with
interventions, the measurement of pain-related fear, and
chronic musculoskeletal pain [59]. A previous systematic
the selection of the reported results were commonly
review showed that PNE can diminish kinesiophobia in
found. The presence of these methodological flaws may
the aforementioned population [60]. However, our re-
influence how the effect size is interpreted. Furthermore,
view found no differences between groups with regard to
some included trials did not report 95% CIs, and the
this intervention. This inconsistency may be associated
sample sizes of some of them were quite small. Both fac-
with differences in the inclusion criteria. Our systematic
tors favor the growth of imprecision, which makes it dif-
review was focused exclusively on fibromyalgia, whereas
ficult to have a precise estimate of the range of the
different chronic musculoskeletal pain conditions, such
observed effect size [66, 67].
as chronic spinal pain and knee osteoarthritis, as well as
chronic fatigue syndrome and fibromyalgia, were in-
cluded in the previous systematic review [60]. Future Research
Fibromyalgia is a condition characterized by high levels The present systematic review has some strengths. First,
of uncertainty related to the illness [61], which may ex- several methodological flaws were detected in the in-
plain why the exclusive application of pain education cluded studies. For example, the presence of serious and
may be insufficient in reducing pain-related fear in people very serious risk of bias or imprecision across the in-
with this condition. This fact is related to another result cluded studies decreased the overall quality of the ran-
found in the present systematic review because PNE did domized clinical trials. Pain-related fear is a relevant
not reduce pain-related anxiety. Finally, no interventions factor in people suffering from chronic pain, and thus, it
were shown to reduce fear-avoidance beliefs and fear of is essential to improve the quality of randomized clinical
pain. Although the present systematic review has trials to draw more robust conclusions about how to
reduce pain-related fear in fibromyalgia. Future research management and electrotherapy were not more effective
should aim to: than controls in decreasing kinesiophobia. Furthermore,
no differences between groups were found for the rest of
1. Increase the precision of the reported findings. This could be
the interventions and pain-related fear constructs (fear-
done through the development of randomized controlled tri-
avoidance beliefs, fear of pain, and pain-related anxiety).
als that enroll larger samples and report precise effect sizes
Nevertheless, a serious risk of bias and a very serious risk
with 95% CIs.
of imprecision were detected across the included trials.
This caused the overall certainty of the evidence to be
2. Reduce the common bias found in randomized controlled tri-
low and very low across interventions. Additionally, the
als. The Consolidated Standards of Reporting Trials
included trials reported insufficient details to permit a
(CONSORT) recommendations are a useful guideline for
full replication of their experimental interventions.
conducting high-quality trials [68].
Therefore, specific clinical implications cannot be derived

3. Deliver complete information about how interventions were from the available evidence. This systematic review has
conducted. The TIDieR checklist may help researchers to in- shown that there are promising interventions that may
clude all the relevant details about their interventions. decrease one specific type of fear in people with fibromy-
algia, e.g., kinesiophobia. However, because of the low
to very low quality of evidence found, a call for action is
needed to improve the quality of randomized clinical tri-
Study Limitations
als that will lead to more definitive information about
First, we intended to meta-analyze data obtained from
the clinical efficacy of interventions in this field.
the included trials. However, several factors, as explained
in the Data Synthesis and Analysis section, hampered
pooling the effect sizes and precluded conducting addi-
Acknowledgment
tional analyses, such as sensitivity analyses and meta-
regression analyses. Second, the low number of included The authors express their gratitude to the Universidad de
trials in each intervention category caused some of the Malaga for its support through the postdoctoral grant
items included in the Synthesis Without Meta-analysis obtained by Dr. Javier Martinez-Calderon.
reporting guidelines to not be followed. For instance,
mean and interquartile range were not used to summarize
the effect sizes. Third, a deviation was made from our re- Authors’ Contributions
view protocol. In our review protocol, we considered All authors have made a substantial scientific contribu-
fear as a global entity. However, we decided to focus the tion to the study in terms of 1) conception and design, or
present systematic review on pain-related fear. This acquisition of data, or analysis and interpretation of
caused some fear constructs to not be considered, such as data, 2) drafting the article or revising it critically for im-
fear of falling. Fourth, the conclusions of the present sys- portant intellectual content, and 3) final approval of the
tematic review should not be extrapolated to chronic version to be published.
musculoskeletal pain conditions other than fibromyalgia.
Clinical Implications Supplementary Data

This systematic review has shown that there are promis- Supplementary Data may be found online at http://pain-
ing interventions (exercise, multicomponent, and psycho- medicine.oxfordjournals.org.
logical interventions) that may decrease one specific type
of fear in people with fibromyalgia, i.e., kinesiophobia.
However, the low to very low quality of the studies calls References
for higher-quality randomized clinical trials in the field 1. Evans CJ, Parthan A, Le K. Economic and humanistic burden of
to establish the clinical efficacy of these interventions. fibromyalgia in the USA. Expert Rev Pharmacoecon Outcomes
Res 2006;6(3):303–14.
2. Lee LK, Ebata N, Hlavacek P, et al. Humanistic and economic
burden of fibromyalgia in Japan. J Pain Res 2016;9:967–78.
Summary and Conclusions
3. Galvez-S anchez CM, Duschek S, Reyes Del Paso GA.
This systematic review aimed to evaluate the effective- Psychological impact of fibromyalgia: Current perspectives.
ness of different interventions in reducing pain-related Psychol Res Behav Manag 2019;12:117–27.
fear in people with fibromyalgia. Twelve randomized 4. Cagnie B, Coppieters I, Denecker S, et al. Central sensitization in
fibromyalgia? A systematic review on structural and functional
clinical trials satisfied the eligibility criteria, including 11
brain MRI. Semin Arthritis Rheum 2014;44(1):68–75.
cohorts with a total sample of 1,441 participants. 5. Clauw DJ. Fibromyalgia: An overview. Am J Med 2009;122
Exercise, multicomponent, and psychological interven- (12):S3–13.
tions seem to be effective in reducing kinesiophobia 6. Arnold LM, Bennett RM, Crofford LJ, et al. AAPT diagnostic
when compared with controls. On the other hand, self- criteria for fibromyalgia. J Pain 2019;20(6):611–28.
7. Vincent A, Benzo RP, Whipple MO, et al. Beyond pain in fibro- 27. Malfliet A, Van Oosterwijck J, Meeus M, et al. Kinesiophobia
myalgia: Insights into the symptom of fatigue. Arthritis Res Ther and maladaptive coping strategies prevent improvements in pain
2013;15(6):221. catastrophizing following pain neuroscience education in fibro-
8. Wu Y-L, Chang L-Y, Lee H-C, Fang S-C, Tsai P-S. Sleep distur- myalgia/chronic fatigue syndrome: An explorative study.
bances in fibromyalgia: A meta-analysis of case-control studies. J Physiother Theory Pract 2017;22(8):653–60.
Psychosom Res 2017;96:89–97. 28. Larsson A, Palstam A, Löfgren M, et al. Pain and fear avoidance
9. Gracely RH, Ceko M, Bushnell MC. Fibromyalgia and depres- partially mediate change in muscle strength during resistance ex-
sion. Pain Res Treat 2012;2012:1–9. ercise in women with fibromyalgia. J Rehabil Med 2017;49
10. Jimenez-Rodrıguez I, Garcia-Leiva JM, Jimenez-Rodriguez BM, (9):744–50.
et al. Suicidal ideation and the risk of suicide in patients with fi- 29. Amer-Cuenca JJ, Pecos-Martın D, Martınez-Merinero P, et al.
bromyalgia: A comparison with non-pain controls and patients How much is needed? Comparison of the effectiveness of differ-
suffering from low-back pain. Neuropsychiatr Dis Treat 2014; ent pain education dosages in patients with fibromyalgia. Pain
10:625–30. Med 2020;21(4):782–93.
11. Macfarlane GJ, Barnish MS, Jones GT. Persons with chronic 30. Friesen LN, Hadjistavropoulos HD, Schneider LH, et al.

widespread pain experience excess mortality: Longitudinal Examination of an internet-delivered cognitive behavioural pain
results from UK Biobank and meta-analysis. Ann Rheum Dis management course for adults with fibromyalgia: A randomized
2017;76(11):1815–22. controlled trial. Pain 2017;158(4):593–604.
12. Arnold LM, Hudson JI, Keck PE, et al. Comorbidity of fibromy- 31. de Medeiros SA, de Almeida Silva HJ, do Nascimento RM, et al.
algia and psychiatric disorders. J Clin Psychiatry 2006;67 Mat Pilates is as effective as aquatic aerobic exercise in treating
(8):1219–25. women with fibromyalgia: A clinical, randomized and blind
13. Ifergane G, Buskila D, Simiseshvely N, Zeev K, Cohen H. trial. Adv Rheumatol 2020;60(1):21.
Prevalence of fibromyalgia syndrome in migraine patients. 32. Higgins JPT, Green S, eds. Cochrane Handbook for Systematic
Cephalalgia 2006;26(4):451–6. Reviews of Interventions, Version 5.1.0 [updated March 2011].
14. Yang T-Y, Chen C-S, Lin C-L, et al. Risk for irritable bowel syn- The Cochrane Collaboration; 2011.
drome in fibromyalgia patients: A national database study. 33. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement
Medicine (Baltimore) 2015;94(10):e616. for reporting systematic reviews and meta-analyses of studies
15. Briones-Vozmediano E, Vives-Cases C, Ronda-Perez E, Gil- that evaluate health care interventions: Explanation and elabora-
Gonzalez D. Patients’ and professionals’ views on managing fi- tion. J Clin Epidemiol 2009;62(10):e1–34–e34.
bromyalgia. Pain Res Manag 2013;18(1):19–24. 34. Haddaway NR, Collins AM, Coughlin D, Kirk S. The role of
16. Quintner J. Why are women with fibromyalgia so stigmatized? Google Scholar in evidence reviews and its applicability to grey
Pain Med 2020;21(5):882–8. literature searching. PLoS One 2015;10(9):e0138237.
17. Sturge-Jacobs M. The experience of living with fibromyalgia: 35. Nicholas M, Vlaeyen JWS, Rief W, et al. The IASP classification
Confronting an invisible disability. Res Theory Nurs Pract 2002; of chronic pain for ICD-11: Chronic primary pain. Pain 2019;
16(1):19–31. 160(1):28–37.
18. Taylor AG, Adelstein KE, Fischer-White TG, Murugesan M, 36. Kwon Y, Lemieux M, McTavish J, Wathen N. Identifying and
Anderson JG. Perspectives on living with fibromyalgia. Glob removing duplicate records from systematic review searches. J
Qual Nurs Res 2016;3:233339361665814. Med Libr Assoc 2015;103(4):184–8.
19. Luque-Suarez A, Martinez-Calderon J, Falla D. Role of kinesio- 37. Sterne JAC, Savovic J, Page MJ, et al. RoB 2: A revised tool for
phobia on pain, disability and quality of life in people suffering assessing risk of bias in randomised trials. BMJ 2019;366:1–8.
from chronic musculoskeletal pain: A systematic review. Br J 38. Büttner F, Winters M, Delahunt E, et al. Identifying the ‘incredi-
Sports Med 2019;53(9):554–9. ble’! Part 2: Spot the difference—A rigorous risk of bias assess-
20. Martinez-Calderon J, Flores-Cortes M, Morales-Asencio JM, ment can alter the main findings of a systematic review. Br J
Luque-Suarez A. Pain-related fear, pain intensity and function in Sports Med 2019;54(13):801–8.
individuals with chronic musculoskeletal pain: A systematic re- 39. Booth J, Moseley GL, Schiltenwolf M, et al. Exercise for chronic
view and meta-analysis. J Pain 2019;20(12):1394–1515. musculoskeletal pain: A biopsychosocial approach.
21. Sanchez AI, Martınez MP, Mir o E, Medina A. Predictors of the Musculoskeletal Care 2017;15(4):413–21.
pain perception and self-efficacy for pain control in patients with 40. Barlow J. How to use education as an intervention in osteoarthri-
fibromyalgia. Span J Psychol 2011;14(1):366–73. tis. Best Pract Res Clin Rheumatol 2001;15(4):545–58.
22. Jones J, Rutledge DN, Jones KD, Matallana L, Rooks DS. Self- 41. McBeth J, Prescott G, Scotland G, et al. Cognitive behavior ther-
assessed physical function levels of women with fibromyalgia: A apy, exercise, or both for treating chronic widespread pain. Arch
national survey. Women’s Health Issues 2008;18(5):406–12. Intern Med 2012;172(1):48–57.
23. Davies KA, Macfarlane GJ, McBeth J, Morriss R, Dickens C. 42. Beasley M, Prescott GJ, Scotland G, et al. Patient-reported
Insecure attachment style is associated with chronic widespread improvements in health are maintained 2 years after completing
pain. Pain 2009;143(3):200–5. a short course of cognitive behaviour therapy, exercise or both
24. Hadlandsmyth K, Dailey DL, Rakel BA, et al. Somatic symptom treatments for chronic widespread pain: Long-term results from
presentations in women with fibromyalgia are differentially asso- the MUSICIAN randomised controlled trial. RMD Open 2015;1
ciated with elevated depression and anxiety. J Health Psychol (1):e000026.
2017;25(6):819–29. 43. McKenzie JB. Chapter 12: Synthesizing and presenting findings
25. Firestone KA, Carson JW, Mist SD, Carson KM, Jones KD. using other methods. In: Higgins JPT, Thomas J, Chandler J, et
Interest in yoga among fibromyalgia patients: An international al., eds. Cochrane Handbook for Systematic Reviews of
internet survey. Int J Yoga Therap 2014;24(1):117–24. Interventions version 60 (updated July 2019). Cochrane; 2019.

26. Russell D, Alvarez Gallardo IC, Wilson I, et al. “Exercise to me Available at: www.training.cochrane.org/handbook.
is a scary word”: Perceptions of fatigue, sleep dysfunction, and 44. Campbell M, McKenzie JE, Sowden A, et al. Synthesis without
exercise in people with fibromyalgia syndrome—a focus group meta-analysis (SWiM) in systematic reviews: Reporting guide-
study. Rheumatol Int 2018;38(3):507–15. line. BMJ 2020;368:l6890.
45. Dailey DL, Vance CGT, Rakel BA, et al. Transcutaneous electri- Cochrane Collaboration; 2011. Available at: www.cochrane-
cal nerve stimulation reduces movement-evoked pain and fa- handbook.org.
tigue: A randomized controlled trial. Arthritis Rheumatol 2020; 57. Guyatt GH, Oxman AD, Vist G, et al. GRADE guidelines: 4.
72(5):824–36. Rating the quality of evidence—Study limitations (risk of bias). J
46. Simister HD, Tkachuk GA, Shay BL, et al. Randomized con- Clin Epidemiol 2011;64(4):407–15.
trolled trial of online acceptance and commitment therapy for fi- 58. Hoffmann TC, Glasziou PP, Boutron I, et al. Better reporting of
bromyalgia. J Pain 2018;19(7):741–53. interventions: Template for intervention description and replica-
47. van Koulil S, Kraaimaat FW, van Lankveld W, et al. Cognitive- tion (TIDieR) checklist and guide. BMJ 2014;348(mar07
behavioral mechanisms in a pain-avoidance and a pain- 3):g1687.
persistence treatment for high-risk fibromyalgia patients. 59. Martinez-Calderon J, Flores-Cortes M, Morales-Asencio JM,
Arthritis Care Res (Hoboken) 2011;63(6):800–7. Luque-Suarez A. Conservative interventions reduce fear in indi-
48. Van Oosterwijck J, Meeus M, Paul L, et al. Pain physiology edu- viduals with chronic low back pain: A systematic review. Arch
cation improves health status and endogenous pain inhibition in Phys Med Rehabil 2020;101(2):329–58.
fibromyalgia: A double-blind randomized controlled trial. Clin J 60. Watson JA, Ryan CG, Cooper L, et al. Pain neuroscience educa-

Pain 2013;29(10):873–82. tion for adults with chronic musculoskeletal pain: A mixed-
49. Carleton RN, Richter AA, Asmundson GJG. Attention modifica- methods systematic review and meta-analysis. J Pain 2019;20
tion in persons with fibromyalgia: A double blind, randomized (10):1140.e1–1140.e22.
clinical trial. Cogn Behav Ther 2011;40(4):279–90. 61. Reich JW, Johnson LM, Zautra AJ, Davis MC. Uncertainty of
50. Carleton RN, Asmundson GJG, Korol SL, et al. Evaluating the illness relationships with mental health and coping processes in
efficacy of an attention modification program for patients with fibromyalgia patients. J Behav Med 2006;29(4):307–16.
fibromyalgia: A randomized controlled trial. Pain 2020;161 62. Holden S, Rathleff MS, Jensen MB, Barton CJ. How can we im-
(3):584–94. plement exercise therapy for patellofemoral pain if we don’t
51. Larsson A, Palstam A, Löfgren M, et al. Resistance exercise know what was prescribed? A systematic review. Br J Sports
improves muscle strength, health status and pain intensity in fi- Med 2018;52(6):385.
bromyalgia—A randomized controlled trial. Arthritis Res Ther 63. Hardeman W, Michie S, Fanshawe T, et al. Fidelity of delivery of
2015;17(1):161. a physical activity intervention: Predictors and consequences.
52. Wilson DB. Practical meta-analysis effect size calculator. Psychol Health 2008;23(1):11–24.
Available at: https://campbellcollaboration.org/research-resour- 64. Carroll C, Patterson M, Wood S, et al. A conceptual framework
ces/effect-size-calculator.html. for implementation fidelity. Implement Sci 2007;2:40.
53. Altman DG, Bland JM. How to obtain the P value from a confi- 65. Turk DC, Adams LM. Using a biopsychosocial perspective in the
dence interval. BMJ 2011;343(aug08 1):d2304. treatment of fibromyalgia patients. Pain Manag 2016;6
54. Cohen J. Statistical Power Analysis for the Behavioral Sciences, (4):357–69.
2nd edition. Hillsdale: NJ: Erlbaum; 1988. 66. Attia A. Why should researchers report to confidence interval in
55. Schünemann HJ, Brozek JGG. GRADE Handbook for Grading modern research? Middle East Fertil Soc J 2005;10(1):78–81.
Quality of Evidence and Strength of Recommendation. GRADE 67. Prel J-B. d, Hommel G, Röhrig B, Blettner M. Confidence inter-
Working Group; 2013. Available at: guidelinedevelopment.org/ val or P-value? Part 4 of a series on evaluation of scientific publi-
handbook. cations. Dtsch Aerzteblatt Online 2009;106(19):335–9.
56. Schünemann HJ, Oxman AD, Vist GE, et al. Chapter 12: 68. Schulz KF, Altman DG, Moher D, for the CONSORT Group.
Interpreting results and drawing conclusions. In: Higgins JPT, CONSORT 2010 Statement: Updated guidelines for reporting
Green S, eds. Cochrane Handbook for Systematic Reviews of parallel group randomised trials. BMJ 2010;340(mar23
Interventions, Version 5.1.0 [updated March 2011]. The 1):c332–702.

Martínez Calderón

Uploaded by

Copyright:

Available Formats

Martínez Calderón

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Martínez Calderón

Uploaded by

Copyright:

Available Formats

Pain Medicine, 22(2), 2021, 481–498

Intervention Therapies to Reduce Pain-Related Fear in Fibromyalgia

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Study registration: Open Science Framework (https://osf.io/8s5z3).

Key words: Chronic Pain; Fibromyalgia; Fear; Review

Introduction online public database (http://doi.org/10.17605/OSF.IO/

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

4. Feasibility and pilot trials.

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

659 studies screened by title and abstract after

62 studies excluded for the

1. Not randomised clinical trials

Eligibility 2. Protocol or feasibility trials (8)

4. No adults with fibromyalgia (1)

5. Pain-related fear only assessed at

6. The full text was unavailable

7. Fear analysed inside a broad

9. Written other language than

Figure 1. PRISMA flow diagram for study selection.

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Sample Characteristics Duration/Contact Fear Characteristics

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

(95% CI)¼4.4 (-3.4

Fear-avoidance beliefs Fear-avoidance beliefs

Fear-avoidance beliefs Fear-avoidance beliefs

about physical ac-

about work: 30.3

about physical ac-

about work: 25.8

tivity: 8.1 (6.9)

Completeness of Intervention Descriptions

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

rationale of the intervention, doses, and materials and

procedures applied in each intervention. On the other

The Overall Certainty of the Evidence: The GRADE

The overall certainty of the available evidence on the ef-

therapy in decreasing kinesiophobia was low because of

the presence of a serious risk of bias and imprecision.

fectiveness of the rest of interventions in reducing kinesi-

pain, was judged to be very low because of a very serious

risk of bias and imprecision. Further details are listed in

Psychological Therapy and Kinesiophobia (Low

was more effective than usual care in reducing kinesio-

(Cohen’s d¼0.63, 95% CI 0.12 to 1.15, P¼0.01; me-

dium effect) [46]. Furthermore, Internet-delivered cogni-

effect) [30]. Finally, telephone-delivered cognitive behav-

randomization) (Cohen’s d¼0.31, 95% CI 0.02 to 0.59,

P¼0.03; small effect) [41] and 24-month follow-up

(Cohen’s d¼0.44, 95% CI 0.14 to 0.74, P¼0.004; small

Table 2. Risk-of-bias assessment

Bias in Bias in the

Downloaded from https://academic.oup.com/painmedicine/article/22/2/481/5912685 by guest on 31 August 2022

H¼high risk of bias; SC¼some concerns of bias; L¼low risk of bias.

Bias in randomization process