Publication bias in the medical literature: A review by a Canadian Research Ethics Board

Dalhousie University and the Capital District Health Authority, Halifax, Nova Scotia, Canada.
Canadian Journal of Anaesthesia (Impact Factor: 2.53). 06/2007; 54(5):380-8. DOI: 10.1007/BF03022661
Source: PubMed


We reviewed the publication record of all protocols submitted to the Capital District Health Authority Research Ethics Board (REB) in Halifax, Nova Scotia, for the period 1995-1996. Because of a heightened awareness of the issue, we hypothesized that there would be less publication bias (a failure to report negative results) and a higher publication rate from completed studies, than previously reported.
Closed studies were identified from the REB database. Publications were identified by the investigators, requests from sponsors, and a literature review. For each publication, we identified authors, title, journal, number of subjects enrolled, and whether or not the publication was a report of a randomized clinical trial. Comparisons were done using a Student's t test, the Chi-square statistic, or Fisher's exact test as appropriate.
From the database of closed studies, 106 remained unpublished, while completed investigations resulted in 84 publications (44% publication rate). The median time to publication was 32.5 months. Publication of statistically significant results occurred in 71/84 trials. Publication of protocols submitted by departments ranged from 91% (anesthesia; 10/11) to 25% [nursing; 2/8 (P<0.05)]. Trials investigating new drugs in Phase 3 or 4 studies were more likely to be published than trials investigating agents in Phase 1 or 2 (P<0.05), and were less likely to be published if sponsored by a pharmaceutical company (P<0.05).
Publication bias continues to be a problem, particularly for early phase investigative studies. Our results suggest that a different approach is required to reduce publication bias. The role that REBs and peer-reviewed journals might play requires further exploration.

1 Follower
21 Reads
  • Source
    • "An independent registration and validating system may help to reduce the bias from both publication process and experimental process. Seeking a different approach is necessary to reduce the impact of this problem (Hall et al 2007). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Publication bias has been around for about 50 years. It has become a concern for almost 20 years in the medical research community. This review briefly summarizes the current status of publication bias, potential sources where bias may arise from, and its common evaluation methods. In the field of translational stroke research, publication bias has long been suspected; however, it has not been addressed with sufficient efforts. Its status has remained the same during the last decade. The author emphasizes the important role that publishers might play in addressing publication bias.
    Journal of Experimental Stroke and Translational Medicine 01/2009; 2(1):16-21. DOI:10.6030/1939-067X-2.1.16
  • Source
    • "A second feature of the LLLT-literature is that publication bias seems to go in a negative direction. This is distinctly different from the drug trials [63,64] where positive results have been found to account for up to 85% of the published trials in single journals [63], although this bias seems to be lesser or absent in high impact journals [64]. Our review suggests that LLLT trials reporting negative results are more likely to be published than trials with positive results. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Recent reviews have indicated that low level level laser therapy (LLLT) is ineffective in lateral elbow tendinopathy (LET) without assessing validity of treatment procedures and doses or the influence of prior steroid injections. Systematic review with meta-analysis, with primary outcome measures of pain relief and/or global improvement and subgroup analyses of methodological quality, wavelengths and treatment procedures. 18 randomised placebo-controlled trials (RCTs) were identified with 13 RCTs (730 patients) meeting the criteria for meta-analysis. 12 RCTs satisfied half or more of the methodological criteria. Publication bias was detected by Egger's graphical test, which showed a negative direction of bias. Ten of the trials included patients with poor prognosis caused by failed steroid injections or other treatment failures, or long symptom duration or severe baseline pain. The weighted mean difference (WMD) for pain relief was 10.2 mm [95% CI: 3.0 to 17.5] and the RR for global improvement was 1.36 [1.16 to 1.60]. Trials which targeted acupuncture points reported negative results, as did trials with wavelengths 820, 830 and 1064 nm. In a subgroup of five trials with 904 nm lasers and one trial with 632 nm wavelength where the lateral elbow tendon insertions were directly irradiated, WMD for pain relief was 17.2 mm [95% CI: 8.5 to 25.9] and 14.0 mm [95% CI: 7.4 to 20.6] respectively, while RR for global pain improvement was only reported for 904 nm at 1.53 [95% CI: 1.28 to 1.83]. LLLT doses in this subgroup ranged between 0.5 and 7.2 Joules. Secondary outcome measures of painfree grip strength, pain pressure threshold, sick leave and follow-up data from 3 to 8 weeks after the end of treatment, showed consistently significant results in favour of the same LLLT subgroup (p < 0.02). No serious side-effects were reported. LLLT administered with optimal doses of 904 nm and possibly 632 nm wavelengths directly to the lateral elbow tendon insertions, seem to offer short-term pain relief and less disability in LET, both alone and in conjunction with an exercise regimen. This finding contradicts the conclusions of previous reviews which failed to assess treatment procedures, wavelengths and optimal doses.
    BMC Musculoskeletal Disorders 02/2008; 9(1):75. DOI:10.1186/1471-2474-9-75 · 1.72 Impact Factor
  • Value in Health 11/2006; 9(6). DOI:10.1016/S1098-3015(10)63525-5 · 3.28 Impact Factor
Show more