Funding, disease area, and internal validity of hepatobiliary randomized clinical trials

Does Type of Sponsorship of Randomized Controlled Trials Influence Treatment Effect Size Estimates in Rehabilitation: A Meta-Epidemiological Study

BACKGROUND

Sponsorship bias could affect research results to inform decision makers when using the results of these trials. The extent to which sponsorship bias affect results in the field of physical therapy has been unexplored in the literature. Therefore, the main aim of this study was to evaluate the influence of sponsorship bias on the treatment effects of randomized controlled trials in physical therapy area.

METHODS

This was a meta-epidemiological study. A random sample of randomized controlled trials included in meta-analyses of physical therapy area were identified. Data extraction including assessments of appropriate influence of funders was conducted independently by two reviewers. To determine the association between biases related to sponsorship biases and effect sizes, a two-level analysis was conducted using a meta-meta-analytic approach.

RESULTS

We analyzed 393 trials included in 43 meta-analyses. The most common sources of sponsorship for this sample of physical therapy trials were government (n = 205, 52%), followed by academic (n = 44, 11%) and industry (n = 39, 10%). The funding was not declared in a high percentage of the trials (n = 85, 22%). The influence of the trial sponsor was assessed as being appropriate in 246 trials (63%) and considered inappropriate/unclear in 147 (37%) of them. We have moderate evidence to say that trials with inappropriate/unclear influence of funders tended to have on average a larger effect size than those with appropriate influence of funding (effect size = 0.15; 95% confidence interval = -0.03 to 0.33).

CONCLUSIONS

Based on our sample of physical therapy trials, it seems that most of the trials are funded by either government and academia and a small percentage are funded by the industry. Treatment effect size estimates were on average 0.15 larger in trials with lack of appropriate influence of funders as compared with trials with appropriate influence of funding. Contrarily to other fields, industry funding was relatively small and their influence perhaps less marked. All these results could be explained by the relative youth of the field and/or the absence of clear industry interests. In front of the call for action by the World Health Organization to strengthen rehabilitation in health systems, these results raise the issue of the need of public funding in the field.

Analysis of the biliostatic potential of two sealants in a standardized porcine model of liver resection

BACKGROUND

Improved resection techniques has decreased mortality rate following liver resections(LRx). Sealants are known as effective adjuncts for haemostasis after LRx. We compared biliostatic effectiveness of two sealants in a standardized porcine model of LRx.

MATERIAL AND METHODS

We accomplished left hemihepatectomy on 27 pigs. The animals were randomized in control group(n = 9) with no sealant and treatment groups (each n = 9), in which resection surfaces were covered with TachoSil^® and TissuFleece^®/Tissucol Duo^®. After 5 days the volume of ascites(ml), bilioma and/or bile leakages and degree of intra-abdominal adhesions were analysed.

RESULTS

Proportion of ascites was lower in TissuFleece/Tissucol Duo^® group. The ascites volume was lower in TachoSil^® group. In sealant groups, increased adhesion specially in the TachoSil^® group was seen. A reduction of the "bilioma rate" was seen in sealant groups, which was significantly lower in TissuFleece^®/Tissucol Duo^® group.

CONCLUSION

In a standardized condition sealants have a good biliostatic effect but with heterogeneous potentials. This property in combination with the cost-benefit analysis should be the focus of future prospective studies.

Bile acid sequestrants for glycemic control in patients with type 2 diabetes: A systematic review with meta-analysis of randomized controlled trials

AIM

To evaluate the effects of bile acid sequestrants (BASs) versus placebo, no intervention or active comparators on glycemic control in type 2 diabetes.

METHODS

Data were retrieved and a systematic review with meta-analyses was performed. We evaluated bias control and subgroup and sensitivity analyses were performed to evaluate heterogeneity and bias.

RESULTS

We included 17 trials with a total of 2950 patients randomized to BASs (colesevelam or colestimide) versus placebo, no intervention, statins or sitagliptin. Random-effects meta-analysis showed that patients randomized to BASs had a lower hemoglobin A_1c at the end of treatment compared with the control group (mean difference-0.55%; 95% confidence interval-0.64 to -0.46). Analysis of trials with low risk of bias in all domains confirmed the findings. Data on adverse events were limited. There were no differences between trials stratified by the control group and no evidence of publication bias or small study effects.

CONCLUSIONS

Our analyses found that BAS treatment improves glycemic control. The size of the effect was clinically relevant and despite limited safety data, our findings support the inclusion of BASs in current diabetes management algorithms for type 2 diabetes.

Systematic characterization of gastrointestinal clinical trials

BACKGROUND

Clinical guidelines are commonly based on inadequate evidence, suggesting deficiencies in the present portfolio of clinical research.

AIMS

To investigate characteristics of clinical trials examining gastrointestinal (GI) diseases registered in ClinicalTrials.gov.

METHODS

A cross-sectional analysis of 13,647 GI trials and 111,535 non-GI trials initiated between January 1997 and September 2013 was performed. Entries were sorted by operational status, purpose, interventions, trial design, and epochs to identify trends and interactions in trial properties.

RESULTS

The global production of GI trials has remained static in recent years and a majority of research efforts are focused on a few diseases. While GI trials are generally produced by highly populated US states and countries, they are also seldom larger than 500 patients. The likelihood of using data monitoring committees, randomization, and double blinding in GI trials has increased over time, though a substantial fraction of GI trials still do not employ rigorous trial designs. While levels of GI trials correlate with disease burden, the explained variance of GI trials by disease burden worldwide is poor.

CONCLUSION

GI trials are chiefly concentrated in few diseases and highly populated regions, exhibit heterogeneous trends and methodologies, and are sensitive to disease burdens, though more so within North America than worldwide.

WITHDRAWN: S-adenosyl-L-methionine for alcoholic liver diseases

The review is withdrawn as it was abandoned and has not been updated since its last edition in 2008. A new team of authors resumed the work on the review, and so far, a major update to the protocol is published. The review is expected to be finalised towards the end of 2016.

The editorial group responsible for this previously published document have withdrawn it from publication.

DEPERROR: Risks of systematic errors in drug and non-drug randomized clinical trials assessing intervention effects in patients with unipolar depression

BACKGROUND

Systematic errors in randomized clinical trials (RCTs) overestimate treatment effects. We systematically assessed the risks of bias in RCTs assessing the effects of drug and non-drug interventions for patients with unipolar depression.

METHODS

We searched bibliographic databases for drug and non-drug RCTs including patients with depression. We assessed the following risk of bias domains: sequence generation, allocation concealment, baseline imbalance, blinding, intention-to-treat analysis, selective outcome reporting, and funding. Risks of bias were compared for drug and non-drug trials and according to year of publication (before 1990; from 1990 to 1999; and 2000 to 2010).

RESULTS

Comparing drug trials (N=775) to non-drug trials (N=73), the proportion of drug trials with low risk of bias seemed superior regarding blinding of participants (p<0.001), blinding of health-care providers (p<0.001), and blinded outcome assessment (p<0.001). Non-drug trials were superior regarding sequence generation (p<0.001), allocation concealment (p=0.002), intention-to-treat analysis (p<0.001), and baseline imbalance (p=0.006). Adequate blinding of data managers (p=0.45), blinding of statisticians (p=0.69), and selective outcome reporting (p=0.55) did not differ. 41.5% of drug trials were funded by for-profit organizations compared to 12.3% of non-drug trials (p<0.001). In drug trials, the risk of bias decreased significantly over time. This did not reach statistical significance in non-drug trials.

LIMITATIONS

This study only included trials published before 2010.

CONCLUSIONS

Included trials were associated with high risks of bias which may distort effect estimates. The risks of bias decreased with time for drug trials.

Effects of glucagon-like peptide-1 receptor agonists on weight loss: systematic review and meta-analyses of randomised controlled trials

OBJECTIVE

To determine whether treatment with agonists of glucagon-like peptide-1 receptor (GLP-1R) result in weight loss in overweight or obese patients with or without type 2 diabetes mellitus.

DESIGN

Systematic review with meta-analyses.

DATA SOURCES

Electronic searches (Cochrane Library, Medline, Embase, and Web of Science) and manual searches (up to May 2011). Review methods Randomised controlled trials of adult participants with a body mass index of 25 or higher; with or without type 2 diabetes mellitus; and who received exenatide twice daily, exenatide once weekly, or liraglutide once daily at clinically relevant doses for at least 20 weeks. Control interventions assessed were placebo, oral antidiabetic drugs, or insulin.

DATA EXTRACTION

Three authors independently extracted data. We used random effects models for the primary meta-analyses. We also did subgroup, sensitivity, regression, and sequential analyses to evaluate sources of intertrial heterogeneity, bias, and the robustness of results after adjusting for multiple testing and random errors.

RESULTS

25 trials were included in the analysis. GLP-1R agonist groups achieved a greater weight loss than control groups (weighted mean difference -2.9 kg, 95% confidence interval -3.6 to -2.2; 21 trials, 6411 participants). We found evidence of intertrial heterogeneity, but no evidence of bias or small study effects in regression analyses. The results were confirmed in sequential analyses. We recorded weight loss in the GLP-1R agonist groups for patients without diabetes (-3.2 kg, -4.3 to -2.1; three trials) as well as patients with diabetes (-2.8 kg, -3.4 to -2.3; 18 trials). In the overall analysis, GLP-1R agonists had beneficial effects on systolic and diastolic blood pressure, plasma concentrations of cholesterol, and glycaemic control, but did not have a significant effect on plasma concentrations of liver enzymes. GLP-1R agonists were associated with nausea, diarrhoea, and vomiting, but not with hypoglycaemia.

CONCLUSIONS

The present review provides evidence that treatment with GLP-1R agonists leads to weight loss in overweight or obese patients with or without type 2 diabetes mellitus.

Assessment of the Methodologic Quality of Medical and Surgical Clinical Trials in Patients with Arthroplasty

Objective.

To assess the methodological quality of randomized controlled trials (RCT) of medical and surgical therapy in patients with arthroplasty.

Methods.

We conducted a Medline database search for all arthroplasty RCT from 1997 and 2006. The quality of the methods of all eligible RCT was assessed by a trained abstractor. We used a checklist of trial quality characteristics, and the overall trial quality was assessed by 3 scales: Jadad (range 0–5), Delphi list (range 0–9), and numeric rating scale (NRS; range 1–10), based on User’s Guides to the Medical Literature.

Results.

A total of 196 articles were included in the analysis; most included hip (n = 81) or knee (n = 80) or both hip/knee arthroplasty (n = 19); 66 (34%) assessed pharmacological treatments, 117 (60%) nonpharmacological treatments, and 13 (7%) both. Mean (SEM) overall quality scores of arthroplasty RCT were low: Jadad score 2.36 (1.4), Delphi list 5.33 (1.6), and NRS score 4.30 (2.6). Multivariable analyses revealed that nonpharmacological intervention RCT had lower odds (odds ratio 0.28–0.39; p = 0.008–0.033) and those with no funding had lower odds (OR 0.28–0.50; p = 0.014–0.119) of being in the highest quartiles of the 3 overall quality scores. In contrast, multicenter RCT had 1.8–4.7 times higher odds of being in highest tertiles of quality scores (p = 0.017–0.185).

Conclusion.

Methodological deficiencies in reporting of hip/knee arthroplasty RCT offer an opportunity for improvement. Type of intervention, number of trial centers, and presence of funding were independently associated with overall trial quality. In future, multicenter RCT (rather than single-center) and modeling protocols of single-center RCT similar in rigor to multicenter RCT may improve the quality of arthroplasty RCT.

Methodological reporting of randomized clinical trials in major gastroenterology and hepatology journals in 2006

To determine the current quality of reporting of randomized clinical trials (RCTs) in the field of gastroenterology and hepatology, we evaluated the methodological reporting of RCTs in six major gastroenterology and hepatology journals. The methodological quality, including generation of the allocation sequence, allocation concealment, double-blinding, and sample size calculation; number of patients; disease area; and funding source was also retrieved from each trial, and the relevant trials were identified by searching MEDLINE in 2006 using a highly sensitive search strategy. The status of reporting the methodological quality of RCTs was descriptively reported. One hundred five trials were included in the final analysis; of these, 81% (85/105) reported adequate generation of the allocation sequence, 61% (64/105) reported adequate allocation concealment, 51% (54/105) were double-blind, and 75% (79/105) reported adequate sample size calculation. The reported methodological quality greatly improved when compared with historical cohorts.

CONCLUSION

This study shows that there was substantial improvement in the reported methodological quality in the major gastroenterology and hepatology journals, but this quality can be further improved.

African HIV/AIDS trials are more likely to report adequate allocation concealment and random generation than North American trials

Background

Adherence to good methodological quality is necessary to minimise bias in randomised conrolled trials (RCTs). Specific trial characteristics are associated with better trial quality, but no studies to date are specific to HIV/AIDS or African trials. We postulated that location may negatively impact on trial quality in regions where resources are scarce.

Methods

1) To compare the methodological quality of all HIV/AIDS RCTs conducted in Africa with a random sample of similar trials conducted in North America; 2) To assess whether location is predictive of trial quality. We searched MEDLINE, EMBASE, CENTRAL and LILACS. Eligible trials were 1) randomized, 2) evaluations of preventive or treatment interventions for HIV/AIDS, 3) reported before 2004, and 4) conducted wholly or partly (if multi-centred) in Africa or North America. We assessed adequacy of random generation, allocation concealment and masking of assessors. Using univariate and multivariate logistic regression analyses we evaluated the association between location (Africa versus North America) and these domains.

Findings

The African search yielded 12,815 records, from which 80 trials were identified. The North American search yielded 13,158 records from which 785 trials were identified and a random sample of 114 selected for analysis. African trials were three times more likely than North American trials to report adequate allocation concealment (OR = 3.24; 95%CI: 1.59 to 6.59; p<0.01) and twice as likely to report adequate generation of the sequence (OR = 2.36; 95%CI: 1.20 to 4.67; p = 0.01), after adjusting for other confounding factors. Additional significant factors positively associated with quality were an a priori sample size power calculation, restricted randomization and inclusion of a flow diagram detailing attrition. We did not detect an association between location and outcome assessor masking.

Conclusions

The higher quality of reporting of methodology in African trials is noteworthy. Most African trials are externally funded, and it is possible that stricter agency requirements when leading trials in other countries and greater experience and training of principal investigators of an international stature, may account for this difference.

Internal and external validity of cluster randomised trials: systematic review of recent trials

OBJECTIVES

To assess aspects of the internal validity of recently published cluster randomised trials and explore the reporting of information useful in assessing the external validity of these trials.

DESIGN

Review of 34 cluster randomised trials in primary care published in 2004 and 2005 in seven journals (British Medical Journal, British Journal of General Practice, Family Practice, Preventive Medicine, Annals of Internal Medicine, Journal of General Internal Medicine, Pediatrics).

DATA SOURCES

National Library of Medicine (Medline) via PubMed.

DATA EXTRACTION

To assess aspects of internal validity we extracted data on appropriateness of sample size calculations and analyses, methods of identifying and recruiting individual participants, and blinding. To explore reporting of information useful in assessing external validity we extracted data on cluster eligibility, cluster inclusion and retention, cluster generalisability, and the feasibility and acceptability of the intervention to health providers in clusters.

RESULTS

21 (62%) trials accounted for clustering in sample size calculations and 30 (88%) in the analysis; about a quarter were potentially biased because of procedures surrounding recruitment and identification of patients; individual participants were blind to allocation status in 19 (56%) and outcome assessors were blind in 15 (44%). In almost half the reports, information relating to generalisability of clusters was poorly reported, and in two fifths there was no information about the feasibility and acceptability of the intervention.

CONCLUSIONS

Cluster randomised trials are essential for evaluating certain types of interventions. Issues affecting their internal validity, such as appropriate sample size calculations and analysis, have been widely disseminated and are now better addressed by researchers. Blinding of those identifying and recruiting patients to allocation status is recommended but is not always carried out. There may be fewer barriers to internal validity in trials in which individual participants are not recruited. External validity seems poorly addressed in many trials, yet is arguably as important as internal validity in judging quality as a basis for healthcare intervention.

Impact of allocation concealment on conclusions drawn from meta-analyses of randomized trials

BACKGROUND

Randomized trials without reported adequate allocation concealment have been shown to overestimate the benefit of experimental interventions. We investigated the robustness of conclusions drawn from meta-analyses to exclusion of such trials.

MATERIAL

Random sample of 38 reviews from The Cochrane Library 2003, issue 2 and 32 other reviews from PubMed accessed in 2002. Eligible reviews presented a binary effect estimate from a meta-analysis of randomized controlled trials as the first statistically significant result that supported a conclusion in favour of one of the interventions.

METHODS

We assessed the methods sections of the trials in each included meta-analysis for adequacy of allocation concealment. We replicated each meta-analysis using the authors' methods but included only trials that had adequate allocation concealment. Conclusions were defined as not supported if our result was not statistically significant.

RESULTS

Thirty-four of the 70 meta-analyses contained a mixture of trials with unclear or inadequate concealment as well as trials with adequate allocation concealment. Four meta-analyses only contained trials with adequate concealment, and 32, only trials with unclear or inadequate concealment. When only trials with adequate concealment were included, 48 of 70 conclusions (69%; 95% confidence interval: 56-79%) lost support. The loss of support mainly reflected loss of power (the total number of patients was reduced by 49%) but also a shift in the point estimate towards a less beneficial effect.

CONCLUSION

Two-thirds of conclusions in favour of one of the interventions were no longer supported if only trials with adequate allocation concealment were included.

Bicyclol for chronic hepatitis C

BACKGROUND

Bicyclol is a novel synthetic 'anti-hepatitis' drug, used in China for chronic hepatitis C. Until now, systematic reviews on bicyclol therapy have not been performed.

OBJECTIVES

To study the benefits and harms of bicyclol for patients with chronic hepatitis C.

SEARCH STRATEGY

We searched The Cochrane Hepato-Biliary Group Controlled Trials Register (July 2005), the Cochrane Central Register of Controlled Trials in The Cochrane Library (Issue 2, 2005), MEDLINE (1994 to July 2005), EMBASE (1994 to July 2005), Science Citation Index Expanded (1994 to July 2005), The Chinese Biomedical Database (1994 to August 2005), Chinese Journals Full-article Database (1994 to August 2005), VIP Chinese Science and Technique Journals Database (1994 to August 2005), and China National Infrastructure (CNKI) (1994 to August 2005). We also contacted manufacturers and researchers in the field.

SELECTION CRITERIA

Randomised clinical trials on bicyclol versus no intervention, placebo, or other interventions in patients with chronic hepatitis C were included, irrespective of blinding, publication status, or language.

DATA COLLECTION AND ANALYSIS

The primary outcome measures were total and liver-related mortality and liver-related morbidity (eg, cirrhosis and carcinoma). Secondary outcome measures were viral response and liver histology.

MAIN RESULTS

The search identified one randomised clinical trial comparing bicyclol with placebo for 39 patients with chronic hepatitis C. The follow-up was three months. In the trial, there was no evidence that bicyclol was superior to placebo for clearance of HCV RNA (RR 3.80, 95% CI 0.47 to 31.0) and anti-HCV antibodies, but bicyclol was associated with statistically significant decrease in alanine aminotransferase activity at 12-th week after treatment (WMD -69 IU/L, 95% CI -115 to -24).

AUTHORS' CONCLUSIONS

Only one randomised clinical trial has examined bicyclol for patients with chronic hepatitis C. This small, short-term trial found no evidence to support or refute its use. Large, randomised, double-blind clinical trials with long-term follow-up are needed to examine the possible benefits and harms associated with bicyclol. Bicyclol can only be recommended for use in randomised trials.

S-adenosyl-L-methionine for alcoholic liver diseases

BACKGROUND

Alcohol is a major cause of liver disease and disrupts methionine and oxidative balances. S-adenosyl-L-methionine (SAMe) acts as a methyl donor for methylation reactions and participates in the synthesis of glutathione, the main cellular antioxidant. Randomised clinical trials have addressed the question whether SAMe may benefit patients with alcoholic liver diseases.

OBJECTIVES

To evaluate the beneficial and harmful effects of SAMe for patients with alcoholic liver diseases.

SEARCH STRATEGY

We searched The Cochrane Hepato-Biliary Group Controlled Trials Register (May 2005), The Cochrane Central Register of Controlled Trials in The Cochrane Library (Issue 2, 2005), MEDLINE (1950 to May 2005), EMBASE (1980 to May 2005), and Science Citation Index Expanded (searched May 2005).

SELECTION CRITERIA

We included randomised clinical trials studying patients with alcoholic liver diseases. Interventions encompassed per oral or parenteral administration of SAMe at any dose versus placebo or no intervention.

DATA COLLECTION AND ANALYSIS

We performed all analyses according to the intention-to-treat method using RevMan Analyses provided by the Cochrane Collaboration. We evaluated the methodological quality of the randomised clinical trials by quality components.

MAIN RESULTS

We identified nine randomised clinical trials including a heterogeneous sample of 434 patients with alcoholic liver diseases. The methodological quality regarding randomisation was generally low, but 8 out of 9 trials were placebo controlled. Only one trial including 123 patients with alcoholic cirrhosis used adequate methodology and reported clearly on all-cause mortality and liver transplantation. We found no significant effects of SAMe on all-cause mortality (relative risks (RR) 0.62, 95% confidence interval (CI) 0.30 to 1.26), liver-related mortality (RR 0.68, 95% CI 0.31 to 1.48), all-cause mortality or liver transplantation (RR 0.55; 95% CI 0.27 to 1.09), or complications (RR 1.35, 95% CI 0.84 to 2.16), but the analysis is based mostly on one trial only. SAMe was not significantly associated with non-serious adverse events (RR 4.92; 95% CI 0.59 to 40.89) and no serious adverse events were reported.

AUTHORS' CONCLUSIONS

We could not find evidence supporting or refuting the use of SAMe for patients with alcoholic liver diseases. We need more long-term, high-quality randomised trials on SAMe for these patients before SAMe may be recommended for clinical practice.

Bias in clinical intervention research

Research on bias in clinical trials may help identify some of the reasons why investigators sometimes reach the wrong conclusions about intervention effects. Several quality components for the assessment of bias control have been suggested, but although they seem intrinsically valid, empirical evidence is needed to evaluate their effects on the extent and direction of bias. This narrative review summarizes the findings of methodological studies on the influence of bias in clinical trials. A number of methodological studies suggest that lack of adequate randomization in published trial reports may be associated with more positive estimates of intervention effects. The influence of double-blinding and follow-up is less clear. Several studies have found a significant association between funding sources and pro-industry conclusions. However, the methodological studies also show that bias is difficult to detect and appraise. The extent of bias in individual trials is unpredictable. A-priori exclusion of trials with certain characteristics is not recommended. Appraising bias control in individual trials is necessary to avoid making incorrect conclusions about intervention effects.

Agreement between reported use of interventions for liver diseases and research evidence in Cochrane systematic reviews

BACKGROUND/AIMS

This study evaluates the agreement between reported use of interventions for patients with liver diseases and research evidence in Cochrane systematic reviews.

METHODS

In July 2002, the Cochrane Hepato-Biliary Group had completed 28 systematic reviews on 36 interventions that were available in Denmark. Based on the reviews, three interventions (n-acetylcysteine for paracetamol overdose, terlipressin for bleeding oesophageal varices, and antibiotics for patients with cirrhosis and gastrointestinal bleeding) with significant beneficial effects on clinical outcomes were classified as 'evidence-based', whereas 19 were classified as 'possibly evidence-based', and 14 as 'not evidence-based'. Questionnaires on reported use and perceived intervention effects were mailed to 108 physicians practising in Danish hospitals. Sixty-six returned their questionnaire.

RESULTS

The proportion of physicians who reported that they never used the three evidence-based interventions varied considerably (2, 62, and 57%, respectively). The perceived intervention effect, duration of clinical experience, employment as head of department, and university hospital employment were significant predictors of more frequent use of evidence-based interventions. Physicians also reported that they used the interventions that were not evidence-based more often if they were employed at a university hospital.

CONCLUSIONS

Considerable disagreements between reported use and research evidence were identified. Additional research on methods to introduce evidence-based medicine in practice seems warranted.

The journal impact factor as a predictor of trial quality and outcomes: cohort study of hepatobiliary randomized clinical trials

OBJECTIVES

To examine the association between the impact factor and characteristics of hepatobiliary randomized clinical trials.

METHODS

A cohort study of 530 hepatobiliary randomized clinical trials was performed. The journal impact factor was extracted from Science Citation Index. For each trial, we extracted the sample size, the quality of randomization and blinding methods, and the statistical significance of the primary outcome measure.

RESULTS

The median sample size was 45 participants (interquartile range 25-88). The allocation sequence generation was adequate in 273 trials (52%). Allocation concealment was adequate in 178 trials (34%). The primary outcome measure was statistically significant in 374 (71%) trials. Nonparametric analyses for trend indicated that the impact factor was significantly associated with the sample size (p < 0.01) and the proportion of trials with adequate allocation sequence generation (p < 0.01) or allocation concealment (p= 0.02). The impact factor was not significantly associated with the study outcome (p= 0.28).

CONCLUSIONS

The present study supports the use of the impact factor as a rough quality indicator. However, even trials in high impact journals may be small or may have inadequate quality. Critical appraisal of individual trials is always necessary, irrespective of the place of publication.

A decade of trials of interferon-alpha for chronic hepatitis C. A meta-regression analysis

The most relevant randomized controlled trials of interferon-alpha (IFN) for naive patients with chronic hepatitis C (CHC) published in a decade, just before appearance of pegylated IFN trials in 2000, were included in this paper. Its purpose is to review the relationship between sustained biochemical response in active versus control group versus usual clinical variables as IFN regimens, cirrhosis, genotype and versus less frequently addressed variables as funding, methodological quality or location of principal author. Meta-analysis estimates of global treatment effect varied according to trial design: group 1=IFN versus placebo/no treatment, 32 RCTs, 2499 pts, OR 9.5 (6.3-14.2); group 2a=comparison of IFN schedules, 43 RCTs, 7454 pts, OR 1.6 (1.4-1.9); group 2b=IFN+other drugs versus standard IFN, 30 RCTs, 4737 pts, OR 2.0 (1.6-2.6). Fixed effects (arm-level) meta-regression on the complete data set (171 arms, 10,580 pts) revealed that sustained response was most likely in experimental arms of IFN+ribavirin or other drugs (OR 2.4), arms using yearly schedule (OR 2.0), trial principal author from Asia (OR 1.7), trial sample size >200 (OR 1.4) and arms enrolling less than 50% of cirrhotics (OR 1.3). Moreover, focus was on some significant interactions too, as the effect of trial's quality interacting to the recorded funding (more benefit if no-profit, less if for-profit) and the effect of trial funding interacting to the location of first author (more benefit if from Asia). Three main effects (experimental arm, cirrhosis, funding) and one interaction (funding*location of principal author) explained 31% of between study variability in a random-effect meta-regression. In a subgroup analysis on a data set including available information on HCV genotype (93 arms, around 7000 pts), meta-regression revealed that genotype 1 or 4 less than 50% per arm and specialistic journal were significant predictors of either biochemical (transaminases) or virological (HCV-RNA) sustained response, in a model including the same main effects identified in the complete data set analysis. Finally, although mostly captured by different IFN regimens along time, heterogeneity of effect in a large set of (not-pegylated) IFN trials was also explained by HCV genotype and variables of quality and reporting, such as trial's principal author from Asia.

Epidemiologic studies of adverse effects of anti-retroviral drugs: how well is statistical power reported

PURPOSE

To determine whether there is a difference in average statistical power between pharmacoepidemiologic studies of anti-retroviral adverse drug effects (ADEs) sponsored by for-profit versus non-profit organizations.

METHODS

We studied all published pharmacoepidemiologic studies of ADEs associated with the 15 anti-retroviral drugs approved through the end of 1999. A priori, the primary outcome was the power of each study to detect a clinically important difference in the risk for an adverse effect among patients exposed to the study drug(s). We could not evaluate this outcome because of the infrequent reporting of power calculations. We instead report the distribution of studies across a 5-tiered measure of adequacy of reporting of statistical power, as well as the sponsorship of these studies.

RESULTS

Of 48 studies meeting our inclusion criteria, only 1 (2%) reported either a completed, a priori power calculation or sufficient details for readers to calculate the power to detect a pre-defined, clinically important effect. Thirty-five studies (73%) reported the minimum information required for sophisticated readers to determine the power to detect an event rate of interest to them; 6 additional studies (13%) reported confidence intervals around at least one summary effect measure and 6 (13%) provided no indication of power or uncertainty. Of the 41 studies for which sponsorship was determined, only 3 (7%) were sponsored by for-profit organizations.

CONCLUSIONS

The poor reporting of statistical power in this sample suggests a need for guidelines to improve the reporting of pharmacoepidemiologic studies of ADEs. Future research is needed to determine whether the observed paucity of industry-sponsored observational studies of anti-retroviral ADEs extends to other clinical areas, and if so, to identify the causes of this phenomenon.