The design and conduct of clinical trials to limit missing data.

This article summarizes recommendations on the design and conduct of clinical trials of a National Research Council study on missing data in clinical trials. Key findings of the study are that (a) substantial missing data is a serious problem that undermines the scientific credibility of causal conclusions from clinical trials; (b) the assumption that analysis methods can compensate for substantial missing data is not justified; hence (c) clinical trial design, including the choice of key causal estimands, the target population, and the length of the study, should include limiting missing data as one of its goals; (d) missing-data procedures should be discussed explicitly in the clinical trial protocol; (e) clinical trial conduct should take steps to limit the extent of missing data; (f) there is no universal method for handling missing data in the analysis of clinical trials - methods should be justified on the plausibility of the underlying scientific assumptions; and (g) when alternative assumptions are plausible, sensitivity analysis should be conducted to assess robustness of findings to these alternatives. This article focuses on the panel's recommendations on the design and conduct of clinical trials to limit missing data. A companion paper addresses the panel's findings on analysis methods.

Structure and composition of the cytoskeleton of nucleated erythrocytes I. The presence of microtubule-associated protein 2 in the marginal band.

The marginal band (MB) of nucleated erythrocytes is composed of a bundle of microtubules that encircles the cell immediately beneath the plasma membrane. When cells are lysed in buffer containing Triton X-100 the MB microtubules remain intact, and the nucleus remains suspended at the cell center by a filamentous network called the trans-MB material that connects the nucleus to the peripheral MB. When these lysed cells are prepared for indirect immunofluorescence by use of an antibody to chick brain microtubule-associated protein 2 (MAP 2), intense staining of the MB results; no staining is evident in the areas occupied by the nucleus or the trans-MB material. Controls demonstrate that the staining is specific, because no staining occurs with fluorescent goat antirabbit serum alone or when nonimmune serum is used as the first antibody. Furthermore, the fluorescence of the MB is not affected by pretreatment of the immune serum with purified tubulin, but staining is prevented by pretreatment of the immune serum with purified MAP 2. To determine which protein component of the MB was responsible for the positive immunofluorescence results, 125I-protein A staining was used after the protein components of the isolated cytoskeleton had been resolved by SDS-polyacrylamide gels. Controls showed that the antiserum could react on SDS gels with MAP 2 from purified chick brain microtubules. The results with the cytoskeletal proteins demonstrated that the antiserum reacted only with a high molecular weight protein having a molecular weight similar, but not identical, to that of chick brain MAP 2. Thus, it is concluded that a protein with antigenic characteristics similar to those of chick brain MAP 2 is a component of the MB. The results are discussed in terms of the possible function of MAP 2 in the MB.

Surgery for nonarteritic anterior ischemic optic neuropathy.

BACKGROUND: Nonarteritic ischemic optic neuropathy is characterized by sudden and painless loss of vision in the eye, accompanied by pallid swelling of the optic disc. No medical therapy has been proven effective in treating this condition. OBJECTIVES: The objective of this review is to assess the safety and efficacy of surgery compared with other treatment or usual care in people with nonarteritic ischemic optic neuropathy. SEARCH STRATEGY: We searched the Cochrane Central Register of Controlled Trials (CENTRAL), which contains the Cochrane Eyes and Vision Group Trials Register, in The Cochrane Library, MEDLINE, EMBASE and the UK National Research Register (NRR). The last search was on 13 July 2005. SELECTION CRITERIA: We included randomized controlled trials. DATA COLLECTION AND ANALYSIS: We obtained full copies of all potentially relevant articles. One randomized trial was eligible for inclusion. One author extracted data which was verified by another author. No synthesis was required. MAIN RESULTS: The one included trial randomized 258 participants and was stopped early. At the time of the 24-month report the follow-up rate was 95.3% for six months and 67.4% for 24 months (174 participants, 89 careful follow up and 85 surgery). There was no evidence of a benefit of surgery on visual acuity. At six months 32.0% of the surgery group had improved visual acuity by three or more lines compared with 42.6% of the careful follow up group (unadjusted risk ratio (RR) 0.75, 95% confidence interval (CI) 0.54 to 1.04). At 24 months 29.4% of the surgery group had improved compared with 31.0% of the careful follow up group (unadjusted RR 0.95, 95% CI 0.60 to 1.49). Participants who underwent surgery had a greater risk of losing three or more lines of vision. At six months 18.9% in the surgery group had worsened compared with 14.8% in the careful follow up group (RR 1.28; 95% CI 0.73 to 2.24). At 24 months 20.0% in the surgery group had worsened compared with 21.8% in the careful follow up group (RR 0.92; 95% CI 0.51 to 1.64). AUTHORS' CONCLUSIONS: Results from the single trial indicate no evidence of a beneficial effect of optic nerve decompression surgery for nonarteritic ischemic optic neuropathy.

Retrospective and prospective identification of unpublished controlled trials: lessons from a survey of obstetricians and pediatricians.

Investigations in which statistically significant differences between treatment groups have not been observed are less likely than others to be reported in scientific journals. In clinical research, this selective suppression of "negative" results may lead to the adoption of ineffective or hazardous treatments. In an attempt to obtain information about unpublished trials in perinatal medicine, letters were sent to 42,000 obstetricians and pediatricians in 18 countries. As a result, we were notified of 395 unpublished randomized trials. Only 18 of the trials had been completed more than 2 years before the survey, a period during which at least 2300 reports of perinatal trials had been published. Of the 395 unpublished trials, 125 had ceased recruitment within the 2 years prior to the survey, 193 were actively recruiting at the time of the survey, and 59 were about to begin recruitment. It was concluded that publication bias will not be addressed successfully by attempts to obtain information about unpublished trials retrospectively. However, since the response rate to our request for details about ongoing and planned trials was good, prospective registration of trials at inception appears to be a feasible approach to reducing publication bias and its adverse consequences. An additional merit of prospective registration of clinical trials is that it should reduce unnecessary duplication (as opposed to necessary replication) in research and promote more effective collaboration.

Publication bias: the problem that won't go away.

Conclusions about the efficacy and safety of medical interventions are based on data presented in the scientific literature. The validity of these conclusions is threatened if publication bias results from investigators or editors making decisions about publishing study results on the basis of the direction or strength of the study findings. This paper reports meta-analyses performed using data from four prospective investigations in which a total of 997 initiated studies were followed to learn of study results, publication status, and reasons for nonpublication. The analysis indicates that there is a positive association between "significant" study results and publication (OR = 2.88; 95% confidence interval [CI] 2.13 to 3.90). When the analysis was restricted to controlled trials (n = 280), an even stronger relationship between "significant" results and publication was observed (OR = 6.15; 95% CI 2.24 to 16.92), with randomized trials (n = 200) apparently no less susceptible to publication bias than controlled trials in general (OR = 8.72; 95% CI 1.91 to 39.81). In every case, failure to publish was investigator-based, and not due to editorial decisions. The results of clinical trials should not be suppressed in this way. Development of registration systems for randomized trials is essential if this problem is to be minimized in future.

Participation in the Ischemic Optic Neuropathy Decompression Trial: sex, race, and age.

BACKGROUND: The Ischemic Optic Neuropathy Decompression Trial (IONDT) is a randomized, single masked, multicenter trial designed to assess the safety and efficacy of optic nerve decompression surgery compared with careful follow-up in patients 50 years or older with non-arteritic anterior ischemic neuropathy (NAION). OBJECTIVES: To examine and evaluate the sex, race, and age distributions of the screened, and subsequently enrolled, IONDT population, especially the proportions of female, minority, and elderly patients, and demographic characteristics of clinical center investigators, the geographical location of IONDT Clinical Centers, and the referral patterns of local physicians. SETTING: Twenty-five U.S. clinical centers. PARTICIPANTS: There were 1,681 referrals to the Clinical Centers; an Eligibility Screening Form providing demographic information was completed for 1,152. FINDINGS: Forty-three percent (495/1,152) of screened cases were women. Seven percent (85/1,152) were minorities: 33 African-Americans, 34 Hispanics, 17 Asians, and 1 Native-American. The average age was 66 +/- 10 years with a range of 22-92 years of age. Of the 1,152 referred patients, 305 were eligible for randomization; 258 (85%) of these agreed to participate in the trial. The demographic makeup of the randomized IONDT patients was similar to that of the screened population. CONCLUSIONS: Women and the elderly are well represented in the IONDT. Because the number of participating minorities was low, we cannot reliably assess their level of participation. Both a low incidence of NAION in minorities and a low referral rate of minorities to clinical trials are plausible explanations for our findings.

How important is publication bias? A synthesis of available data.

It has long been recognized that investigators frequently fail to report their research findings (Dickersin, 1990). Chalmers (1990) has suggested that this failure represents scientific misconduct since volunteers who consent to participate in research, and agencies that provide funding support for investigations, do so with the understanding that the work will make a contribution to knowledge. Clearly, knowledge that is not disseminated is not making a "contribution". This failure to publish is not only inappropriate scientific conduct, it also influences the information available for interpretation by the scientific community. Of course, if research is left randomly unpublished, there is less information available, but that information is unbiased. We now have solid evidence that failure to publish is not a random event; rather, publication is dramatically influenced by the direction and strength of research findings (Dickersin et al., 1987, 1992; Dickersin & Min, 1993; Easterbrook et al., 1991; Simes, 1986). This tendency of editors and reviewers to accept manuscripts submitted by investigators based on the strength and direction of the research findings is termed "publication bias". The problem has been under discussion for many years and has recently been studied directly in medicine and public health. This article will review the major evidence available regarding publication bias and will suggest measures for overcoming the problem.

Effectiveness of the 40 adolescent AIDS-risk reduction interventions: a quantitative review.

OBJECTIVE: (1) To review evaluations of Acquired Immunodeficiency Syndrome (AIDS) prevention interventions targeting adolescents to determine whether these efforts have been effective in reducing risk behaviors. (2) To examine the relation between intervention design issues and outcomes measures. METHOD: A systematic review of five electronic databases and hand-searches of 11 journals, were undertaken for articles published from January 1983 through November 1995 that reported evaluations of adolescent AIDS risk-reduction interventions conducted in the United States. Outcomes examined include: improved attitudes about AIDS risk and protective behaviors, increased intention to abstain from sexual intercourse or to use condoms, and reduced sexual risk behaviors. RESULTS: Forty studies that met preestablished inclusion criteria were found. For each outcome assessed, a majority of studies found a positive intervention impact (88% of studies assessing changes in knowledge; 58% changes in attitude, 60% changes in intention to use condoms, 73% in condom use, and 64% in decreasing number of sexual partners). Interventions that demonstrated an increase in intention to use a condom were significantly more likely to be theory-based than those that did not show any significant changes in intention (100% vs. 0%, p = .048). Interventions that increased condom use and decreased the number of sexual partners were longer in duration than those that did not improve these outcomes. CONCLUSIONS: AIDS risk reduction interventions can be effective in improving knowledge, attitudes, and behavioral intentions and in reducing risk practices. The positive relationship between improved outcomes and several elements of intervention design underscores the need for increased focus on intervention design in future studies.

Surgery for nonarteritic anterior ischemic optic neuropathy.

BACKGROUND: Nonarteritic ischemic optic neuropathy is characterized by sudden and painless loss of vision in one eye, accompanied by pallid swelling of the optic disc. Although various medical interventions, such as corticosteroids and phenytoin sodium, have been used to treat nonarteritic ischemic optic neuropathy, no therapy has been proven effective. OBJECTIVES: The objective of this review is to assess the safety and efficacy of surgical treatment compared with other treatment or usual care in people with nonarteritic ischemic optic neuropathy. SEARCH STRATEGY: We searched the Cochrane Controlled Trials Register - Central and MEDLINE. The most recent searches were performed in December 1997. SELECTION CRITERIA: We included randomized trials comparing surgery to no surgery in people with nonarteritic ischemic optic neuropathy. DATA COLLECTION AND ANALYSIS: We obtained full copies of all potentially relevant articles. Only one article described a randomized trial of surgery and it was eligible for inclusion. No formal assessment of quality was done. One reviewer extracted data. No synthesis was required, as there was only one trial. MAIN RESULTS: The one trial identified randomized 258 patients. The only published report with outcomes data for that trial presents preliminary results from 244 patients who had achieved six months of follow-up at the time of the report. Participants assigned to surgery did no better than participants assigned to careful follow-up regarding improved visual acuity of three or more lines of vision at six months: 32.6% of the surgery group improved compared with 42.7% of the careful follow-up group. The adjusted odds ratio (OR), adjusted for baseline visual acuity and diabetes, comparing the two groups for three or more lines improvement was 0.74 (95% confidence interval (CI) 0.39 to 1. 38) (surgery group improvement was worse than careful follow-up). In addition, participants receiving surgery had a significantly greater risk of losing three or more lines of vision at six months: 23.9% in the surgery group worsened compared with 12.4% in the careful follow-up group. The six-month adjusted OR comparing the two groups for loss of three or more lines of vision was 1.96 (95% CI 0.87 to 4.41). Spontaneous improvement of three or more lines of vision was observed in 42.7% of participants in the careful follow-up group. REVIEWER'S CONCLUSIONS: Results from the Ischemic Optic Neuropathy Decompression Trial indicate that optic nerve decompression surgery for nonarteritic ischemic optic neuropathy is not effective.

Systematic literature review for clinical practice guideline development.

The purpose of this paper was to evaluate the quality and scope of the published literature on functional impairment due to cataract in adults as reviewed for the Agency for Health Care Policy and Research Clinical Practice Guideline. We examined the method of literature retrieved and analysis performed in the course of development of literature-based recommendations for the guideline panel. To collect data, we reviewed the process of literature acquisition and identification and the quality assessments made by reviewers of 14 individual topics composed of 77 issues related to the guideline. We collated this information to provide an assessment of the quality and scope of the relevant literature. Less than 4% (310) of the approximately 8,000 articles initially identified as potentially relevant to the guideline were ultimately used. The majority covered three topics (surgery and complication, 100; Nd:YAG capsulotomy, 77; and potential vision testing, 40). Three other topics--indications for surgery, preoperative medical evaluation, and rehabilitation--were devoid of articles meeting inclusion criteria. For 43 issues, there was no identifiable relevant literature. With few exceptions, the quality of the literature was rated fair to poor owing to major flaws in experimental design. Case series (256 reports) of one type or another accounted for the majority of the included literature. There were 17 random controlled trials. This review revealed a sparse and generally low-quality literature relevant to the management of functional impairment due to cataract, despite a relatively large data base in reputable peer-reviewed journals.

Identifying relevant studies for systematic reviews.

OBJECTIVE: To examine the sensitivity and precision of Medline searching for randomised clinical trials. DESIGN: Comparison of results of Medline searches to a "gold standard" of known randomised clinical trials in ophthalmology published in 1988; systematic review (meta-analysis) of results of similar, but separate, studies from many fields of medicine. POPULATIONS: Randomised clinical trials published in 1988 in journals indexed in Medline, and those not indexed in Medline and identified by hand search, comprised the gold standard. Gold standards for the other studies combined in the meta-analysis were based on: randomised clinical trials published in any journal, whether indexed in Medline or not; those published in any journal indexed in Medline; or those published in a selected group of journals indexed in Medline. MAIN OUTCOME MEASURE: Sensitivity (proportion of the total number of known randomised clinical trials identified by the search) and precision (proportion of publications retrieved by Medline that were actually randomised clinical trials) were calculated for each study and combined to obtain weighted means. Searches producing the "best" sensitivity were used for sensitivity and precision estimates when multiple searches were performed. RESULTS: The sensitivity of searching for ophthalmology randomised clinical trials published in 1988 was 82%, when the gold standard was for any journal, 87% for any journal indexed in Medline, and 88% for selected journals indexed in Medline. Weighted means for sensitivity across all studies were 51%, 77%, and 63%, respectively. The weighted mean for precision was 8% (median 32.5%). Most searchers seemed not to use freetext subject terms and truncation of those terms. CONCLUSION: Although the indexing terms available for searching Medline for randomised clinical trials have improved, sensitivity still remains unsatisfactory. A mechanism is needed to "'register" known trials, preferably by retrospective tagging of Medline entries, and incorporating trials published before 1966 and in journals not indexed by Medline into the system.

Hysterectomy, endometrial ablation and Mirena® for heavy menstrual bleeding: a systematic review of clinical effectiveness and cost-effectiveness analysis.

OBJECTIVE: The aim of this project was to determine the clinical effectiveness and cost-effectiveness of hysterectomy, first- and second-generation endometrial ablation (EA), and Mirena® (Bayer Healthcare Pharmaceuticals, Pittsburgh, PA, USA) for the treatment of heavy menstrual bleeding. DESIGN: Individual patient data (IPD) meta-analysis of existing randomised controlled trials to determine the short- to medium-term effects of hysterectomy, EA and Mirena. A population-based retrospective cohort study based on record linkage to investigate the long-term effects of ablative techniques and hysterectomy in terms of failure rates and complications. Cost-effectiveness analysis of hysterectomy versus first- and second-generation ablative techniques and Mirena. SETTING: Data from women treated for heavy menstrual bleeding were obtained from national and international trials. Scottish national data were obtained from the Scottish Information Services Division. PARTICIPANTS: Women who were undergoing treatment for heavy menstrual bleeding were included. INTERVENTIONS: Hysterectomy, first- and second-generation EA, and Mirena. MAIN OUTCOME MEASURES: Satisfaction, recurrence of symptoms, further surgery and costs. RESULTS: Data from randomised trials indicated that at 12 months more women were dissatisfied with first-generation EA than hysterectomy [odds ratio (OR): 2.46, 95% confidence interval (CI) 1.54 to 3.93; p = 0.0002), but hospital stay [WMD (weighted mean difference) 3.0 days, 95% CI 2.9 to 3.1 days; p < 0.00001] and time to resumption of normal activities (WMD 5.2 days, 95% CI 4.7 to 5.7 days; p < 0.00001) were longer for hysterectomy. Unsatisfactory outcomes associated with first- and second-generation techniques were comparable [12.2% (123/1006) vs 10.6% (110/1034); OR 1.20, 95% CI 0.88 to 1.62; p = 0.2). Rates of dissatisfaction with Mirena and second-generation EA were similar [18.1% (17/94) vs 22.5% (23/102); OR 0.76, 95% CI 0.38 to 1.53; p = 0.4]. Indirect estimates suggested that hysterectomy was also preferable to second-generation EA (OR 2.32, 95% CI 1.27 to 4.24; p = 0.006) in terms of patient dissatisfaction. The evidence to suggest that hysterectomy is preferable to Mirena was weaker (OR 2.22, 95% CI 0.94 to 5.29; p = 0.07). In women treated by EA or hysterectomy and followed up for a median [interquartile range (IQR)] duration of 6.2 (2.7-10.8) and 11.6 (7.9-14.8) years, respectively, 962/11,299 (8.5%) women originally treated by EA underwent further gynaecological surgery. While the risk of adnexal surgery was similar in both groups [adjusted hazards ratio 0.80 (95% CI 0.56 to 1.15)], women who had undergone ablation were less likely to need pelvic floor repair [adjusted hazards ratio 0.62 (95% CI 0.50 to 0.77)] and tension-free vaginal tape surgery for stress urinary incontinence [adjusted hazards ratio 0.55 (95% CI 0.41 to 0.74)]. Abdominal hysterectomy led to a lower chance of pelvic floor repair surgery [hazards ratio 0.54 (95% CI 0.45 to 0.64)] than vaginal hysterectomy. The incidence of endometrial cancer following EA was 0.02%. Hysterectomy was the most cost-effective treatment. It dominated first-generation EA and, although more expensive, produced more quality-adjusted life-years (QALYs) than second-generation EA and Mirena. The incremental cost-effectiveness ratios for hysterectomy compared with Mirena and hysterectomy compared with second-generation ablation were £1440 per additional QALY and £970 per additional QALY, respectively. CONCLUSIONS: Despite longer hospital stay and time to resumption of normal activities, more women were satisfied after hysterectomy than after EA. The few data available suggest that Mirena is potentially cheaper and more effective than first-generation ablation techniques, with rates of satisfaction that are similar to second-generation techniques. Owing to a paucity of trials, there is limited evidence to suggest that hysterectomy is preferable to Mirena. The risk of pelvic floor surgery is higher in women treated by hysterectomy than by ablation. Although the most cost-effective strategy, hysterectomy may not be considered an initial option owing to its invasive nature and higher risk of complications. Future research should focus on evaluation of the clinical effectivesness and cost-effectiveness of the best second-generation EA technique under local anaesthetic versus Mirena and types of hysterectomy such as laparoscopic supracervical hysterectomy versus conventional hysterectomy and second-generation EA. FUNDING: The National Institute for Health Research Health Technology Assessment programme.

Hysterectomy, endometrial destruction, and levonorgestrel releasing intrauterine system (Mirena) for heavy menstrual bleeding: systematic review and meta-analysis of data from individual patients.

OBJECTIVE: To evaluate the relative effectiveness of hysterectomy, endometrial destruction (both "first generation" hysteroscopic and "second generation" non-hysteroscopic techniques), and the levonorgestrel releasing intrauterine system (Mirena) in the treatment of heavy menstrual bleeding. DESIGN: Meta-analysis of data from individual patients, with direct and indirect comparisons made on the primary outcome measure of patients' dissatisfaction. DATA SOURCES: Data were sought from the 30 randomised controlled trials identified after a comprehensive search of the Cochrane Library, Medline, Embase, and CINAHL databases, reference lists, and contact with experts. Raw data were available from 2814 women randomised into 17 trials (seven trials including 1359 women for first v second generation endometrial destruction; six trials including 1042 women for hysterectomy v first generation endometrial destruction; one trial including 236 women for hysterectomy v Mirena; three trials including 177 women for second generation endometrial destruction v Mirena). Eligibility criteria for selecting studies Randomised controlled trials comparing hysterectomy, first and second generation endometrial destruction, and Mirena for women with heavy menstrual bleeding unresponsive to other medical treatment. RESULTS: At around 12 months, more women were dissatisfied with outcome with first generation hysteroscopic techniques than with hysterectomy (13% v 5%; odds ratio 2.46, 95% confidence interval 1.54 to 3.9, P<0.001), but hospital stay (weighted mean difference 3.0 days, 2.9 to 3.1 days, P<0.001) and time to resumption of normal activities (5.2 days, 4.7 to 5.7 days, P<0.001) were longer for hysterectomy. Unsatisfactory outcomes were comparable with first and second generation techniques (odds ratio 1.2, 0.9 to 1.6, P=0.2), although second generation techniques were quicker (weighted mean difference 14.5 minutes, 13.7 to 15.3 minutes, P<0.001) and women recovered sooner (0.48 days, 0.20 to 0.75 days, P<0.001), with fewer procedural complications. Indirect comparison suggested more unsatisfactory outcomes with second generation techniques than with hysterectomy (11% v 5%; odds ratio 2.3, 1.3 to 4.2, P=0.006). Similar estimates were seen when Mirena was indirectly compared with hysterectomy (17% v 5%; odds ratio 2.2, 0.9 to 5.3, P=0.07), although this comparison lacked power because of the limited amount of data available for analysis. CONCLUSIONS: More women are dissatisfied after endometrial destruction than after hysterectomy. Dissatisfaction rates are low after all treatments, and hysterectomy is associated with increased length of stay in hospital and a longer recovery period. Definitive evidence on effectiveness of Mirena compared with more invasive procedures is lacking.

Report from the panel on the Case for Registers of Clinical Trials at the Eighth Annual Meeting of the Society for Clinical Trials.

A plenary session at the Eighth Annual Meeting of the Society for Clinical Trials addressed "The Case for Registers of Clinical Trials." There are a number of reasons for having registers of clinical trials: to promote collaboration and communication among investigators regarding new and ongoing clinical trials, to provide a basis for methodologic research, and to facilitate meta-analysis by the availability of a system of trial identification that is independent of the published literature. The experience gleaned by those involved with two existing trial registers, the International Committee on Thrombosis and Haemostasis Registry and the Oxford Database of Perinatal Trials, can be used to provide insight into the issues generic to trial registration. The time perspective to be used, inclusion and exclusion criteria, the practicability of comprehensive prospective registration, and funding are central considerations for these and other registers.

The existence of publication bias and risk factors for its occurrence.

Publication bias is the tendency on the parts of investigators, reviewers, and editors to submit or accept manuscripts for publication based on the direction or strength of the study findings. Much of what has been learned about publication bias comes from the social sciences, less from the field of medicine. In medicine, three studies have provided direct evidence for this bias. Prevention of publication bias is important both from the scientific perspective (complete dissemination of knowledge) and from the perspective of those who combine results from a number of similar studies (meta-analysis). If treatment decisions are based on the published literature, then the literature must include all available data that is of acceptable quality. Currently, obtaining information regarding all studies undertaken in a given field is difficult, even impossible. Registration of clinical trials, and perhaps other types of studies, is the direction in which the scientific community should move.

NIH clinical trials and publication bias.

OBJECTIVE: To investigate the association between trial characteristics, findings, and publication. The major factor hypothesized to be associated with publication was "significant" results, which included both statistically significant results and results assessed by the investigators to be qualitatively significant, when statistical testing was not done. Other factors hypothesized to have a possible association with publication were funding institute, funding mechanism (grant versus contract versus intramural), multicenter status, use of comparison groups, large sample size, type of control (parallel versus nonparallel), use of randomization and masking, type of analysis (by treatment received versus by treatment assigned), and investigator sex and rank. DESIGN: Follow-up, by 1988 interview with the principal investigator or surrogate, of all clinical trials funded by the National Institutes of Health (NIH) in 1979, to learn of trial results and publication status. POPULATION: Two hundred ninety-three NIH trials, funded in 1979. MAIN OUTCOME MEASURE: Publication of clinical trial results. RESULTS: Of the 198 clinical trials completed by 1988, 93% had been published. Trials with "significant" results were more likely to be published than those showing "nonsignificant" results (adjusted odds ratio [OR] = 12.30; 95% confidence interval [CI], 2.54 to 60.00). No other factor was positively associated with publication. Most unpublished trials remained so because investigators thought the results were "not interesting" or they "did not have enough time" (42.8%). Metaanalysis using data from this and 3 similar studies provided a combined unadjusted OR of 2.88 (95% CI, 2.13 to 3.89) for the association between significant results and publication. CONCLUSIONS: Even when the overall publication rate is high, such as for trials funded by the NIH, publication bias remains a significant problem. Given the importance of trials and their utility in evaluating medical treatments, especially within the context of metaanalysis, it is clear that we need more reliable systems for maintaining information about initiated studies. Trial registers represent such a system but must receive increased financial support to succeed.

Publication of clinical trials: accountability and accessibility.

Publication of findings from clinical trials is a necessary step in the research continuum, to provide a record of the work done, convey information to the community, and support translation of research into clinical practice. Systematic reviews of randomized controlled trials are now widely regarded as the highest level of evidence in determining the effect of an intervention on an outcome. They largely depend on internationally accessible, published reports of all trials undertaken. Investigators and their institutions or organizations have responsibility for reporting their clinical trials accurately and completely, including disclosure of potential conflicts of interest. To ensure evidence-based health care, issues relating to accessibility and accountability of clinical trial results require immediate action.