scispace - formally typeset
Search or ask a question

Showing papers by "Peter C Gøtzsche published in 2004"


Journal ArticleDOI
26 May 2004-JAMA
TL;DR: The reporting of trial outcomes is not only frequently incomplete but also biased and inconsistent with protocols and Published articles, as well as reviews that incorporate them, may therefore be unreliable and overestimate the benefits of an intervention.
Abstract: ContextSelective reporting of outcomes within published studies based on the nature or direction of their results has been widely suspected, but direct evidence of such bias is currently limited to case reports.ObjectiveTo study empirically the extent and nature of outcome reporting bias in a cohort of randomized trials.DesignCohort study using protocols and published reports of randomized trials approved by the Scientific-Ethical Committees for Copenhagen and Frederiksberg, Denmark, in 1994-1995. The number and characteristics of reported and unreported trial outcomes were recorded from protocols, journal articles, and a survey of trialists. An outcome was considered incompletely reported if insufficient data were presented in the published articles for meta-analysis. Odds ratios relating the completeness of outcome reporting to statistical significance were calculated for each trial and then pooled to provide an overall estimate of bias. Protocols and published articles were also compared to identify discrepancies in primary outcomes.Main Outcome MeasuresCompleteness of reporting of efficacy and harm outcomes and of statistically significant vs nonsignificant outcomes; consistency between primary outcomes defined in the most recent protocols and those defined in published articles.ResultsOne hundred two trials with 122 published journal articles and 3736 outcomes were identified. Overall, 50% of efficacy and 65% of harm outcomes per trial were incompletely reported. Statistically significant outcomes had a higher odds of being fully reported compared with nonsignificant outcomes for both efficacy (pooled odds ratio, 2.4; 95% confidence interval [CI], 1.4-4.0) and harm (pooled odds ratio, 4.7; 95% CI, 1.8-12.0) data. In comparing published articles with protocols, 62% of trials had at least 1 primary outcome that was changed, introduced, or omitted. Eighty-six percent of survey responders (42/49) denied the existence of unreported outcomes despite clear evidence to the contrary.ConclusionsThe reporting of trial outcomes is not only frequently incomplete but also biased and inconsistent with protocols. Published articles, as well as reviews that incorporate them, may therefore be unreliable and overestimate the benefits of an intervention. To ensure transparency, planned trials should be registered and protocols should be made publicly available prior to trial completion.

1,638 citations


Journal ArticleDOI
TL;DR: This work searched MEDLINE, EMBASE, Web of Science, and the Cochrane Library using a wide array of terms related to harms and identified pertinent evidence and made recommendations on the appropriate reporting of harms in RCTs.
Abstract: In response to overwhelming evidence and the consequences of poor-quality reporting of randomized, controlled trials (RCTs), many medical journals and editorial groups have now endorsed the CONSORT (Consolidated Standards of Reporting Trials) statement, a 22-item checklist and flow diagram Because CONSORT primarily aimed at improving the quality of reporting of efficacy, only 1 checklist item specifically addressed the reporting of safety Considerable evidence suggests that reporting of harms-related data from RCTs also needs improvement Members of the CONSORT Group, including journal editors and scientists, met in Montebello, Quebec, Canada, in May 2003 to address this problem The result is the following document: the standard CONSORT checklist with 10 new recommendations about reporting harms-related issues, accompanying explanation, and examples to highlight specific aspects of proper reporting We hope that this document, in conjunction with other CONSORT-related materials (http://wwwconsort-statementorg), will help authors improve their reporting of harms-related data from RCTs Better reporting will help readers critically appraise and interpret trial results Journals can support this goal by revising Instructions to Authors so that they refer authors to this document

1,279 citations


Journal ArticleDOI
TL;DR: A systematic review of 114 randomized trials that compared placebo‐treated patients with untreated patients could not confirm that placebo interventions induce powerful effects.
Abstract: . Background. It is widely believed that placebo interventions induce powerful effects. We could not confirm this in a systematic review of 114 randomized trials that compared placebo-treated with untreated patients. Aim. To study whether a new sample of trials would reproduce our earlier findings, and to update the review. Methods. Systematic review of trials that were published since our last search (or not previously identified), and of all available trials. Results. Data was available in 42 out of 52 new trials (3212 patients). The results were similar to our previous findings. The updated review summarizes data from 156 trials (11 737 patients). We found no statistically significant pooled effect in 38 trials with binary outcomes, relative risk 0.95 (95% confidence interval 0.89–1.01). The effect on continuous outcomes decreased with increasing sample size, and there was considerable variation in effect also between large trials; the effect estimates should therefore be interpreted cautiously. If this bias is disregarded, the pooled standardized mean difference in 118 trials with continuous outcomes was −0.24 (−0.31 to −0.17). For trials with patient-reported outcomes the effect was −0.30 (−0.38 to −0.21), but only −0.10 (−0.20 to 0.01) for trials with observer-reported outcomes. Of 10 clinical conditions investigated in three trials or more, placebo had a statistically significant pooled effect only on pain or phobia on continuous scales. Conclusion. We found no evidence of a generally large effect of placebo interventions. A possible small effect on patient-reported continuous outcomes, especially pain, could not be clearly distinguished from bias.

335 citations


Journal ArticleDOI
TL;DR: Because high-quality trials failed to demonstrate a reduction in mortality, polyclonal immunoglobulin should not be used for treatment of sepsis except in randomized clinical trials.
Abstract: Randomized trials of adjunctive treatment of bacterial sepsis with polyclonal immunoglobulin show conflicting results. We performed a systematic review and a meta-analysis of the results of randomized trials that compared reductions in mortality rates in patient groups treated with polyclonal immunoglobulin versus either placebo or no treatment in addition to conventional treatment. High-quality trials had adequate concealment of allocation, were double-blinded and placebo-controlled, and made data available for intention-to-treat analyses. Twenty trials were included. Meta-analysis of all trials showed a relative risk of death with immunoglobulin treatment of 0.77 (95% confidence interval [CI], 0.68-0.88). High-quality trials (involving a total of 763 patients, 255 of whom died) showed a relative risk of 1.02 (95% CI, 0.84-1.24), whereas other trials (involving a total of 948 patients, 292 of whom died) showed a relative risk of 0.61 (95% CI, 0.50-0.73). Because high-quality trials failed to demonstrate a reduction in mortality, polyclonal immunoglobulin should not be used for treatment of sepsis except in randomized clinical trials.

128 citations


Journal ArticleDOI
15 Jan 2004-BMJ
TL;DR: Information provided by professional advocacy groups and governmental organisations is information poor and severely biased in favour of screening, and few websites live up to accepted standards for informed consent.
Abstract: Objective To investigate whether information on mammographic screening presented on websites by interest groups is balanced, is independent of source of funding, and reflects recent findings. Design Cross sectional study using a checklist with 17 information items. Setting 27 websites in Scandinavian and English speaking countries. Results The 13 sites from advocacy groups and the 11 from governmental institutions all recommended mammographic screening, whereas the three from consumer organisations questioned screening (P = 0.0007). All the advocacy groups accepted industry funding, apparently without restrictions. In contrast the three consumer organisations acknowledged the risk of bias related to industry funding, and two of them did not accept such funding at all. Advocacy groups and governmental organisations favoured information items that shed positive light on screening. The major harms of screening, overdiagnosis and overtreatment, were mentioned by only four of these groups, but by all three sites from consumer organisations (P = 0.02). In addition, the chosen information was often misleading or erroneous. The selection of information items for websites did not reflect recent findings, apart from the consumer sites, which were much more balanced and comprehensive than other sites (median of 9 information items v3 items, P = 0.03). Conclusions The information material provided by professional advocacy groups and governmental organisations is information poor and severely biased in favour of screening. Few websites live up to accepted standards for informed consent such as those stated in the General Medical Council9s guidelines.

121 citations


Journal ArticleDOI
TL;DR: It is shown that CF patients who are treated intensively have lower antibody responses and longer survival after acquisition of chronic P. aeruginosa lung infection.
Abstract: We studied the effects of increasingly intensive treatment regimens on anti-pseudomonal antibody response and survival in five successive cohorts of a total of 157 Danish cystic fibrosis patients after they had acquired chronic P. aeruginosa lung infection. The time periods were 1971–1975 (N = 21), 1976–1980 (N = 64), 1981–1986 (N = 27), 1987–1993 (N = 26), and 1994–2000 (N = 19). During this 30-year period, we introduced elective 2-week courses of chemotherapy every third month in all chronically infected patients, early aggressive treatment with inhalation of colistin and oral ciprofloxacin for 3 months whenever P. aeruginosa was cultured in sputum from noncolonized patients, and inhalation of recombinant human dornase alfa. There was a significant correlation between the calendar year when chronic P. aeruginosa infection was acquired and the subsequent increase in the level of precipitins (P < 0.00001). The median number of precipitins increased by 5 per year in the oldest calendar year cohort, and 1 per year in the youngest. The median age of onset of chronic P. aeruginosa increased from 9.3 years from 1981–1986 to 13.8 years from 1987–2000. Survival after acquisition of chronic P. aeruginosa lung infection improved with time (P = 0.008). Our study shows that CF patients who are treated intensively have lower antibody responses and longer survival after acquisition of chronic P. aeruginosa lung infection. Pediatr Pulmonol. 2004; 37:427–432. © 2004 Wiely-Liss, Inc.

80 citations


Journal ArticleDOI
TL;DR: People with rheumatoid arthritis and the researchers in the study did prefer non-steroidal anti-inflammatory drugs more than acetaminophen/paracetamol, and there is a need for a large trial with appropriate randomisation, double-blinding, and with explicit methods to measure and analyse pain and adverse effects.
Abstract: Background Nonsteroidal anti-inflammatory drugs (NSAIDs) are usually preferred for simple analgesics such as paracetamol for rheumatoid arthritis. It is not clear, however, whether the trade-offs between benefits and harms of NSAIDs are preferable to those of paracetamol (paracetamol is also called acetaminophen). Objectives To compare the benefits and harms of paracetamol with NSAIDs in patients with rheumatoid arthritis. Search methods PubMed and EMBASE databases were searched up until August 2007. Reference lists of identified articles were also searched. Selection criteria Randomised double-blind studies comparing paracetamol with an NSAID. Data collection and analysis Decisions on inclusion of trials and data extraction were performed by the two authors independently. Main results Four cross-over studies, published between 1968 and 1982, involving 121 patients, and four different NSAIDs were included. The generation of the allocation sequence and the use of methods to conceal the allocation were not described in any of the studies. The studies were double-blind but it was not clear whether the blinding was effective. Methods for collecting adverse effects were not described. The NSAIDs were preferred more often than paracetamol by the patients or the investigator. In the largest trial, 20 out of 54 patients (37%) preferred ibuprofen and 7 out of 54 (13%) paracetamol. Investigators preference (as established by joint tenderness, grip strength and joint circumference) was 17 out of 35 for diclofenac versus 5 out of 35 for paracetamol in another trial. However, because of the weaknesses in the trials, no firm conclusion can be drawn. Authors' conclusions When considering the trade off between the benefits and harms of non-steroidal anti-inflammatory drugs and paracetamol/acetaminophen, it is not known whether one is better than the other for rheumatoid arthritis. But people with rheumatoid arthritis and the researchers in the study did prefer non-steroidal anti-inflammatory drugs more than acetaminophen/paracetamol. There is a need for a large trial, with appropriate randomisation, double-blinding, test of the success of the blinding, and with explicit methods to measure and analyse pain and adverse effects.

62 citations


Journal ArticleDOI
TL;DR: FPR claim that the critique of the randomized screening trials has little merit; that there is no reason to believe that the Canadian study was of better quality than the New York Health Insurance Plan study or the Two-County study; and that the prior consensus on mammography was correct.
Abstract: claim that our critique of the randomized screening trials has little merit; that there is no reason to believe that the Canadian study was of better quality than the New York Health Insurance Plan (HIP) study or the Two-County study; and that the prior consensus on mammography was correct. However, their review suffers from erroneous assumptions and biased statistical analyses, and their quotations are often selective and misleading. In my discussion of the issues, I will follow when possible the sequence of arguments used by FPR. FPR claim in their abstract that early detection leads to less invasive therapy. This could have been true, if the only effect of screening had been to detect the same tumours earlier that are detected later if women are not screened. FPR naively believe that screening does just that, i.e. does not lead to overdiagnosis. They note, for example, that the incidence of breast cancers in the New York HIP study is the same in the control group as in the screening group 5–7 years after screening started (their table 2), as they expected. I will explain under the HIP study below why this argument is faulty. The level of overdiagnosis can be studied reliably in the trials from Canada and Malmö which did not differentially exclude women with prior breast cancer after randomization and did not introduce early, systematic screening of the whole control group. 2–4 There was an overdiagnosis of 30% 5,6 which corresponds closely to the 31% excess surgery we have previously described 2,3 (Table 1). A similar result was seen for the trials that screened the whole control group when only cancers before this screen were included 7–9 (Table 2). The excess surgery rate was 20% for mastectomies. 2,3 We have discussed in detail why it is likely that even today, there would be about 20% more mastectomies when women are screened than if they are not screened. 3 In Southeast Netherlands, for example, when screening was introduced from 1990 to 1998, the number of women who underwent breast-conserving surgery increased by 71%, and the number of women who underwent mastectomy increased by 84%. 10 If the study had included carcinoma in situ there would have been even more mastectomies. 3,11 A study from Italy claimed that screening had not led to an increase in mastectomies, 12 but this study had no control group and the premises for the study …

58 citations


Journal ArticleDOI
TL;DR: The balance between possible benefits and harms is delicate, which underlines the need for honest information to women also on harms and individual decision making rather than blanket recommendations.
Abstract: The issue of breast cancer screening is worthy of continued scrutiny and scientific and public debate [1]. The quality of the scientific evidence is variable, and new evidence appears all the time. Furthermore, the effect of breast cancer screening is uncertain [2], and screening leads to overdiagnosis and overtreatment [3,4]. Hence, the balance between possible benefits and harms is delicate, which underlines the need for honest information to women also on harms and individual decision making rather than blanket recommendations [5].

12 citations




Journal ArticleDOI
25 Mar 2004-BMJ
TL;DR: As Manaszewicz writes, the language and the framing of information items can influence the way a message is perceived (see also the box of examples).
Abstract: EDITOR—As Manaszewicz writes, the language and the framing of information items can influence the way a message is perceived (see also our box of examples).1 Regarding our use of the term harms, we are of course not “the only researchers of note to resort to such language.” For example, this term is …