Evidence Table 2Quality assessment of fibromyalgia trials

Internal Validity
Author, Year
Randomization adequate?Allocation concealment adequate?Groups similar at baseline?Eligibility criteria specified?Outcome assessors masked?Care provider masked?Patient masked?Intention-to-treat (ITT) analysisMaintenance of comparable groupsAcceptable levels of crossovers, adherence, and contamination?Acceptable levels of overall attrition and between-group differences in attrition?Quality Rating
Anderberg, 2000Unclear, “randomization was made”Unclear, “patients were given consecutive codes”Unclear, only reported comparability on depressive and pain symptomsYesUnclear, described as double-blindYesYesYesUnclearNR, NR, NRYes (12.5%)
Arnold, 2002UnclearUnclearUnclear; more married in fluoxetine group (77% vs 50%, P= 0.06)YesUnclear, described as double-blindYesYesNo, excluded 15%UnclearNR, NR, NRNo=38%
Yes=37% for fluoxetine, 40% for placebo
Arnold, 2004YesUnclearYesYesYes (double blind)Yes (double blind)YesYesYesYesNoFair
Arnold, 2005UnclearNRYesYesUnclear (double blind)Unclear (double blind)UnclearYesYesUnclearNoFair
Arnold, 2007NRNRNo
Drug group had significantly lower average pain interference score & higher SF-36 Bodily pain score
YesNRImplied - double- blind, placebo controlled designImplied - double-blind, placebo- controlled designYesYesNR, NR, NRYes, YesFair
Arnold, 2008YesNRYesYesYes (implied double blind)Yes (implied double blind)YesYes, LOCF
5/750 excluded from analysis=0.6%
YesUnclear, Unclear, UnclearYes, YesFair
Arnold, 2010YesYesYesYesYesYesYesYesYesNRNoFair
Ataoglu, 1997UnclearUnclearUnclear, baseline characteristics only reported for analyzed group (90%)YesUnclear, blinding NRUnclear, blinding NRUnclear, blinding NRNo, excluded 10%YesUnclear, Unclear, UnclearOverall: Yes=10%
Between-groups: paroxetine=6%, amitriptyline=15%
Bennett, 1988UnclearUnclearUnclearYesUnclear, described as double-blindYesYesYesYesUnclear, Unclear, UnclearOverall: No (47%)
Between-group: cyclobenzaprine=35%, placebo=60%, P< 0.05
Branco, 2010UnclearNRYesYesYes (double blind)Yes (double blind)YesYes, LOCF. 1 excluded from efficacty analysis due to missed baseline data and 7 from safety (4 due to practice concerns at 1 center and 3 as did not receive any study med)YesYesNoFair
Carette, 1986UnclearNRNRYesUnclear (double blind)Unclear (double blind)YesNoNo (placebo had longer duration of fibromyalgia)YesYesFair
Carette, 1994YesUnclearUnclear; median duration of fibromyalgia lower in cyclobenzaprine group (36 months) compared to other groups (60 months), but difference not statistically significantYesUnclear, described as double-blindYesYesNo, excluded 24/208 (11%)UnclearNR, Yes, YesNo (25%)
No; amitriptyline=16.7%, cyclobenzaprine=29.3%, placebo=33.3%
Carette, 1995YesUnclearUnclearYesUnclear, described as double-blindYesYesNo, excluded 2/22 (9%)UnclearUnclear, Yes, UnclearYes, YesFair
Chappell, 2008YesNRYesYesYes (double blind)Yes (double blind)UnclearYesYesYesNoFair
Clauw, 2008YesYesYesYesYesYesYesYes (LOCF)YesNRNoFair
Crofford, 2005; Arnold, 2007YesNRYesYesNRImplied - double blind, placebo- controlled designImplied - double blind, placebo- controlled designYesYesUnclear, Unclear, UnclearYes, YesFair
Crofford, 2008Unclear, described as telerandomizationNRYesYesNRImplied - double blind, placebo- controlled designImplied - double blind, placebo- controlled designYesYesYes, Unclear, UnclearOverall=Yes
Between-groups=No, Pregabalin
300mg/day:52%, 450mg/day: 67%, 600mg/day: 63%, placebo: 81%
Ginsberg, 1996UnclearUnclearUnclear, excluded 5 (11%) who were lost to follow-upYesUnclear, described as double-blindYesYesNo, excluded 5/51 (11%)YesNR, Yes, YesYes, YesFair
Giordano, 1999Unclear; “separated into 2 groups according to a randomization list”UnclearUnclear, data NRYesUnclear, described as single-blindUnclear, described as single-blindUnclear, described as single-blindYesUnclearNR, NR, NRNo=27.5%
No=15% for paroxetine and 40% for placebo
Goldenberg, 1986UnclearUnclearUnclear; data NR, but statement of “no significant differences with respect to race, duration of fibromyalgia symptoms, prevalence of sleep disturbances, or morning tirednes”, “neither the tender point score nor any other outcome measure differed significant between groups at study onset”YesYesYesYesUnclearYes, excluded 3%NR, Yes, NRYes, YesFair
Goldenberg, 1996YesYesUnclear, crossover study but characteristics only reported for overall groupYesUnclear, described as double-blindYesYesNo, excluded 29% to 39%, depending on outcomeUnclearNR, NR, NRNo=39%
No=4 while receiving fluoxetine, 1 while receiving amitriptyline, 5 while receiving both, and 1 with placebo
Hannonen, 1998YesYesYesYesYes (implied double blind)Tes (implied double blind)YesYesUnclearYes, Yes, YesOverall - NO (29%)
Between grous - Yes (moclobemid 30%, amitriptyline 23%, Placebo 33%)
Heymann, 2001YesUnclearUnclear, no statistical differences, but fewer
Caucasians in nortriptyline group (55%) vs amitriptyline and placebo groups (65% in both) and duration of illness NR
YesUnclear, described as double-blindYesYesNo, excluded 12/118 (10%)UnclearNR, NR, NRYes (14%)
No; placebo=17.5%, amitriptyline=7.5%, nortriptyline=5.3%
Mease, 2008UnclearNRYesYesYes (implied double blind)Yes (implied double blind)Yes (implied double blind)Yes, LOCF
3/751 excluded from analysis=0.4%
YesUnclear, Unclear, UnclearOverall=Yes
Between-groups=No; 41.6%, 450mg/day: 33.9%, 300mg/day: 33.5%, placebo 31.6%, difference between groups (600mg/day and placebo): 10%, p value between groups p=0.044
Mease, 2009UnclearNRYesYesUnclearUnclearYesYes (LOCF)YesYesNoFair
Norregaard, 1995UnclearUnclearYesYesUnclear, described as double-blindYesYesYes, included 42/43 (98%)YesNR, NR, NRNo (10/43=23%)
Unclear; attrition not stratified by treatment group
Patkar, 2007; Pae, 2009YesYesYesYesUnclear, described as double-blindYesYesYesUnclearNR, NR, NRNo=26%
No=34% for paroxetine, 17% for placebo
Pfizer, 2008UnclearNRReported as similar but data NRYesYes (implied double blind)Yes (implied double blind)YesYes - those randomized and took study medNRNRYesFair
Quimby, 1989UnclearUnclearUnclear, statement of "nonsignificant differences", but data NRYesUnclear, described as double-blindYesYesUnclear, numbers analyzed
UnclearUnclear, Unclear, UnclearOverall: Yes (11%)
Between-groups: Yes
Reynolds, 1991UnclearUnclearUnclear; NR based on order of randomizationYesUnclear, described as double-blindUnclear, described as double-blindUnclear, described as double-blindUnclear, numbers analyzed
UnclearUnclear, Unclear, UnclearOverall: No (25%)
Between-groups: Yes
Russell, 2008; Hunter 2009YesNRYesYesUnclear (double blind)Unclear (double blind)UnclearYesYesUnclearNoFair
Scudds, 1989UnclearUnclearUnclear, crossover study but characteristics only reported for overall groupYesUnclear, described as double-blindYesYesNo, excluded 8%UnclearUnclear, Unclear, UnclearOverall: Yes=8%
Between-groups: Yes
Vitton, 2004; Gendreau, 2005YesYesYes except for MDDYesYesYesYesYesUnclearNRNoFair
Wolfe, 1994YesUnclearNo; more high school graduates (90.5% vs 61.9%, P= 0.03) and longer disease duration (16.1 vs 9.6 years; P= 0.05) in fluoxetine groupYesUnclear, described as double-blindUnclear, described as double-blindUnclear, described as double-blindNo; excluded 18/42 (43%)UnclearUnclear, Unclear, UnclearOverall=No (43%)
Between-groups=No (fluoxetine=29%, placebo=57%)

From: Evidence Tables

Cover of Drug Class Review: Drugs for Fibromyalgia
Drug Class Review: Drugs for Fibromyalgia: Final Original Report [Internet].
Smith B, Peterson K, Fu R, et al.
Portland (OR): Oregon Health & Science University; 2011 Apr.
Copyright © 2011, Oregon Health & Science University.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.