Proportion Of Clinical Trials Research Articles

Current situation and reflection of clinical trials in the field of burn medicine in China

Clinical trials are of great significance in formulating guidelines or consensus for the diagnosis and treatment of burns. However, the innate advantages of clinical trials in the field of burn medicine in China have not been translated into evidence-based medical data. Our research group counted the literature published in major academic journals in the field of burn medicine at home and abroad from 2010 to 2020 in Chinese and English databases, and found that the number and proportion of clinical trials in the field of burn medicine published in Chinese journals were generally lower than those in English journals. Moreover, the number and proportion of clinical trials in the field of burn medicine published in Chinese journals were lower than those in the field of critical care medicine. On this basis, our research group statistically analyzed the registration status of clinical trials in the field of burn medicine from 2010 to 2020, and found that the registration volume and completion volume of clinical trials in the field of burn medicine in China were not only lower than those in the field of burn medicine in the United States, but also lower than those in the field of critical care medicine in China. The reasons for insufficient clinical trials in the field of burn medicine may be the decrease in the incidence of burns, the decrease in the number of burn specialists and their low income, and the lack of burn research talents. It is necessary for China to integrate the advantages of clinical resources in the field of burn medicine, strengthen clinical trial research, and improve the discourse power in the international community.

State of use of Electronic Data Capture (EDC) tools in randomized controlled trials in India

ObjectivesElectronic data capture (EDC) tools can improve data quality gathered in clinical trials and may be more cost-effective. However, limited data is available on the adoption of EDC tools in randomized controlled trials (RCT) conducted in India. The objective of the current study was to determine the EDC adoption rate in RCTs in India. MethodsWe invited investigators of registered RCTs in India to an online survey. The questionnaire included questions on the use of EDC and features available. An EDC sophistication level (range 1 - 6) was computed. Respondents were also asked about barriers to the implementation of EDC. The EDC adoption rate (EAR) was defined as the proportion of clinical trials where EDC with a sophistication level of 2 or more was used. Multivariable logistic regression was used to identify factors that predicted EDC adoption. Results321 responses were received for 400 unique trials (participation rate of 21%), with an EAR of 27.5% (95% confidence intervals : 23.4% to 32.1%, n = 110). The number of sites influenced EDC adoption (odds ratio : 1.26, 95% CI : 1.12 - 1.47, p = 0.001) on multivariable analysis. EAR did not increase over time. The key barriers identified for not using an EDC were lack of technical support (170, 63.0%) and software cost (132, 48.9%). ConclusionsThe survey shows a low EAR in randomized trials registered in India. The barriers identified in the survey would need systematic solutions to improve the EAR in the future.

Abstract 463: Highly accurate machine learning assessment of immune-related pathologic response criteria (irPRC) scoring in patients with non-small cell lung carcinoma (NSCLC) treated with neoadjuvant anti-PD-1-based therapies

Abstract Pathological complete response (no residual viable tumor, RVT) and/or major pathologic response (≤10% RVT) are now primary or secondary endpoints for a large proportion of clinical trials studying neoadjuvant immunotherapeutic regimens. We previously developed a scoring system for assessing pathologic response after immunotherapy, termed irPRC (Cottrell et al. Ann Oncol 2018). By these criteria, %RVT is assessed by dividing RVT by the sum of the surface area on the slide composed of RVT + necrosis + regression bed– the latter feature is where the tumor used to be and is characterized by fibroinflammatory stroma that is distinct from tumoral stroma. We have previously reported high inter-observer reproducibility for pathologic response assessment following immunotherapy. However, these assessments involve performing evaluations that are currently outside the scope of routine surgical pathology training and may be time-consuming. To date, these assessments have primarily been performed by academic pathologists who have seen the largest number of these cases as a part of clinical trials. A machine learning (ML)-powered assessment of irPRC would allow for faster, standardized evaluation and expanded access to patients treated outside of large academic centers. We trained a supervised convolutional neural network to assess pathologic response using irPRC on n=92 H&E-stained slides from patients with advanced, resectable NSCLC treated with neoadjuvant anti-PD-1 +/- anti-CTLA-4 at a single institution. The ML algorithm was trained based on ground-truth manual annotations by pathologists on whole slide digital scans and tested using leave-one-out cross validation. Each of ~830,000 image tiles was classified into one of four classes: tumor, necrosis, immune-mediated regression, or background lung tissue. Receiver operating curves showed that the algorithm exhibited high accuracy for predicting the various tissue classes with an area under the curve of 0.95, 0.96, 0.90, and 0.90 for the four classes, respectively. %RVT was calculated by dividing the surface area of RVT by total tumor bed surface area (RVT + necrosis + regression). There was a strong positive correlation between the machine assessed RVT and the human assessed RVT at both the slide level and case level (aggregate %RVT based on surface area from all slides for a given patient), Pearson’s r=0.95 and r=0.99, respectively. Here, we demonstrate that a ML algorithm performs as well as an experienced pathologist assessment in scoring pathologic response. These findings will need to be validated in larger studies. Additionally, the association of pathologic response with longer term patient outcomes will be evaluated as survival data matures to determine whether pathologic response is a robust surrogate of survival. Citation Format: Julie E. Stein, Vinay Pulim, Tricia R. Cottrell, Patrick M. Forde, Janis M. Taube. Highly accurate machine learning assessment of immune-related pathologic response criteria (irPRC) scoring in patients with non-small cell lung carcinoma (NSCLC) treated with neoadjuvant anti-PD-1-based therapies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 463.

Four decades of hand microsurgery in Iran: A bibliographic analysis.

The aim of this study was to perform a bibliometric analysis to assess the number of articles published by Iranian researchers in the field of hand and microsurgery over the last four decades. An online search was conducted using 685 keywords in the abstract/title sections of articles, including carpal tunnel syndrome, wrist fractures, nerve injury and repair, skin flap and graft in the hand, congenital disorders in the hand and forearm, tumor in the hand and wrist, and infection in the hand and wrist. From February 1976 to May 2021, EndNote software version 8.1 was used to search articles in PubMed and Scopus databases. Articles in which at least one of the authors was affiliated with Iran were chosen. The name of the original institution, field of study, total number of publications, type of study, collaboration rate of Iranian hand surgeons for each year, and annual sharing of Iranian articles in journals with the highest cite scores in the field of hand and microsurgery were all examined in the present study. The total number of publications in the field of hand and microsurgery was 632 (an average of 11 papers per year). Most of the Iranian hand and microsurgery papers were from the capital city, Tehran (38.09%). There was an increasing trend in the number of publications over the years, most which were about carpal tunnel syndrome (21.5%), tendon (9.8%), and nerve (9.6%). In total, 59.6% were descriptive articles, whereas the proportion of clinical trials was relatively small (22.3%). Iranian hand microsurgeons have continued to increase their scientific output in hand microsurgery and related fields over the last 40 years. The quality of Iranian hand microsurgeons' scientific output, both within and outside their discipline, has greatly improved; however, they still have a long way to go before becoming a hub of science. IV, retrospective study without control group.

A Database of Drug Repurposing Clinical Trials in Oncology.

Drug repurposing is a strategy that seeks new medical treatments from existing licensed medications rather than from de novo development of new molecules. The rationale for this approach is predicated in part on the perceived advantages of using existing data on safety and toxicity, the ready availability of such medicines and the potential for lowered drug costs (Pantziarka et al., 2014; Bertolini et al., 2015). In particular, the use of non-cancer drugs as new cancer treatments is becoming an increasingly attractive proposition. While bibliometric data shows an exponential increase in the peer reviewed literature in this area, much of it remains focused on drug candidate identification and pre-clinical studies (Pantziarka et al., 2020). There has been a growth of computational support for candidate identification in recent years, with a focus on algorithms and database support to enhance disease-target-drug analysis (Peyvandipour et al., 2018; Zhu et al., 2018; Tanoli et al., 2021). A notable contribution to the field is the PRISM (profiling relative inhibition simultaneously in mixtures) system from the Broad Institute. This open access resource records the growth inhibitory activity of thousands of compounds against more than 500 cancer cell lines (Corsello et al., 2020). The importance of regulatory approvals, intellectual property rights and other non-clinical factors in repurposing have also been highlighted and discussed, particularly by Verbaanderd et al. (2019) and by Begley et al. (2021). To date there has not been a comprehensive analysis of clinical trials in repurposing in oncology. The answers to key questions regarding the proportion of clinical trials by drug, by phase, by geographic location or cancer indication remain unclear. This work describes the methodology for the creation of a curated database of drug repurposing clinical trials in oncology. The database, which we have called the ReDO_Trials_DB, is available as an online, open access database with basic search, filtering and download functionality. Definitions of drug repurposing, sometimes also called drug repositioning, vary widely and can encompass multiple drug development strategies from the further development of previously shelved compounds to the use of licensed cancer drugs in new cancer types to the exploration of non-cancer drugs as new cancer therapeutics (Pushpakom et al., 2019; Pantziarka et al., 2020). In this work we are focused specifically on the use of licensed non-cancer drugs as potential cancer therapeutics. To this end the list of drugs classed as repurposing candidates is derived from the ReDO_DB (https://www.anticancerfund.org/en/redo-db), our previously published drug repurposing database (Pantziarka et al., 2018). To recap, drugs are included in the ReDO_DB if they meet the following criteria: • Licensed by one or more national/international medicines regulatory agency (e.g., FDA, EMA, MHRA etc) • Are not licensed for a cancer indication, although drugs used for symptomatic or diagnostic uses in cancer can be included (e.g., anti-emetics) if there is evidence of anticancer activity as well • Have published evidence of anti-cancer activity, including in vitro, in vivo, case reports, observational studies or clinical trials In this paper we present the ReDO_Trials_DB, the methodology employed to generate and maintain it and a first description of the number and characteristics of the included trials.

Immune effector cell therapies in oncology: A systematic analysis and forecast from ClinicalTrials.gov.

e14516 Background: IEC are arguably the most promising therapy in oncology, and FDA approvals for CAR T cells are a testament to their real potential. This has led to a great growth of clinical trials (CT), making the evaluation of their associated data increasingly challenging. To facilitate this process, we performed a systematic review of ClinicalTrials.gov focused on IEC therapies in oncology. We performed an analysis of the CT data and forecast trends for 2025. Methods: CT registries were retrieved from ClinicalTrials.gov with a systematic search query. We included CT registered between 1993-2020, which used IEC-based therapies in oncology. Statistical analysis using descriptive and inferential methods allowed us to identify trends and establish forecasts. Results: We identified 938 registrations of IEC-CTs in oncology, and 51% of those were active. The most common IEC type was CAR T (51%), followed by NK (15%), TCR T (8%), TIL (8%), and CIK (3%). 69% of IEC-CTs were aimed at a specific target vs. 31% that lack that specificity. The number of annual CAR T-CTs continue to increase, and we forecast 320 registrations during 2025. Since 2018, the number of CTs using unmodified T cells and TIL has increased (largely due to metastatic melanoma studies). NK cells CTs represent 8% of all registrations, and by 2025 we anticipate that >150 NK-based CTs will be registered per year, a number similar to CAR T-CTs registered during 2020. IEC-CTs based on allogeneic sources represent 18% of all registrations and are expected to grow 2.5 fold in the next 5 years. In solid tumors, the most common IEC-CTs are based on Non-CAR studies, while in hem-malignancies 76% of CTs are CAR T. Only one solid tumor, CAR T-CT, has reached phase 2/3, while 7 CIK-CTs were used to target various organ tumors. Conclusions: IEC-CTs continue to grow exponentially and represent an active field of clinical research in oncology. Allogeneic and NK-CTs increase rapidly and are among the most promising IEC-CTs. A rising proportion of CTs in solid tumors are using CIK and TIL rather than CAR T-cells, which are more popular for treating hem-malignancies. This is a direct reflection of the challenges associated with CAR T infiltration and persistence in solid tumors. Our analyses indicate that annual IEC-CTs registrations will double by 2025. We anticipate the increase in cellular therapy options will demand an organized response, including adequate logistical planning and policy implementation in response to the landscape changes in clinical oncology during the upcoming years.[Table: see text]

Changes in clinical trials of endocrine disorder and metabolism and nutrition disorder drugs in mainland China over 2010-2019.

With the improvements in relevant policies, laws, and regulations regarding drug clinical trials in China, the quantity and quality of drug clinical trials have gradually improved, and the development prospects of drug clinical trials for endocrine disorders and metabolism and nutrition disorders are promising. Based on information from the clinical trials from the online drug clinical trial registration platform of the National Medical Products Administration, we aimed to review and evaluate the development of clinical trials of drugs for endocrine disorders and metabolism and nutrition disorders in mainland China from 2010 to 2019, as well as the trends over time. A total of 861 trials were carried out on 254 types of drugs for endocrine disorders and metabolism and nutrition disorders, among which 531 (61.67%) involved endocrine disorders, and 330 (38.33%) addressed metabolism and nutrition disorders. The annual number of clinical trials has been increasing gradually, with a significant increase in 2017. Among them, the proportion of clinical trials with Chinese epidemiological characteristics was relatively large (Wu, Annual Report on Development Health Management and Health Industry in China, 2018). The largest number of trials were for diabetes drugs (55.63%), followed by trials of drugs for hyperlipidemia (19.4%) and those for hyperuricemia (7.9%). It was found that the geographical area of the leading units also showed obvious unevenness according to the analysis of the test unit data. Based on the statistics and evaluation of the data, comprehensive information is provided to support the cooperation of global pharmaceutical R&D companies and research units in China and the development of international multicenter clinical trials in China. This work additionally provides clinical trial units with a self‐evaluation of scientific research competitiveness and hospital development strategies. At the same time, it provides a reference with basic data for sponsors and stakeholders in these trials to determine their development strategy goals.

Sample Size Requirements for Calibrated Approximate Credible Intervals for Proportions in Clinical Trials.

In Bayesian analysis of clinical trials data, credible intervals are widely used for inference on unknown parameters of interest, such as treatment effects or differences in treatments effects. Highest Posterior Density (HPD) sets are often used because they guarantee the shortest length. In most of standard problems, closed-form expressions for exact HPD intervals do not exist, but they are available for intervals based on the normal approximation of the posterior distribution. For small sample sizes, approximate intervals may be not calibrated in terms of posterior probability, but for increasing sample sizes their posterior probability tends to the correct credible level and they become closer and closer to exact sets. The article proposes a predictive analysis to select appropriate sample sizes needed to have approximate intervals calibrated at a pre-specified level. Examples are given for interval estimation of proportions and log-odds.

Brain metastasis as exclusion criteria in clinical trials involving extensive-stage small cell lung cancer.

The American Society of Clinical Oncology and Friends of Cancer Research submitted recommendations to the FDA to reduce barriers in clinical trial participation. They proposed the removal of several specific exclusion criteria, including brain metastasis. Clinical trials involving small cell lung cancer (SCLC) have varying exclusion criteria regarding brain metastasis. We completed an online search of clinicaltrials.gov for the query "SCLC, extensive stage." The trials were classified into a group of strict exclusion, allowed only if treated, allowed without treatment, or undefined. Relationships between status of brain metastasis in exclusion criteria and study characteristics (trial status, trial design, sponsor, location, and treatment groups) were investigated by Chi-squared test. The trends of exclusion status were investigated by a comparison against the variable time. Of the 204 eligible trials, 32 strictly excluded any form or history of CNS metastases, 129 allowed patients that are undergoing or have undergone CNS-specific therapy, 9 allowed patients without any CNS-specific therapy, and 34 did not mention any criteria involving CNS metastases. Studies conducted outside the United States and with single systemic therapy were associated with strict exclusion of brain metastasis (p = 0.026 and 0.039, respectively). The proportion of clinical trials with strict exclusion has remained around 15% for the past few decades. Non-US and single systemic therapy studies are more commonly associated with strict exclusion of brain metastasis in ES-SCLC trials. The strict exclusion of brain metastases in clinical trials has remained relatively constant for the past few decades.

Clinical evaluation of marketed orthodontic products: are researchers behind the times? A meta-epidemiological study

BackgroundThe role of marketing and industry in the treatment decisions of orthodontists has received increasing attention in recent years with clinical research typically undertaken subsequent to established use of these devices and often failing to confirm the promise of manufacturers’ claims. This meta-epidemiological study was undertaken to assess the proportion of clinical trials in orthodontics evaluating commercially marketed products and to evaluate the direction of the results of these studies.MethodsElectronic searching was undertaken to identify randomized controlled trials (RCTs) published over a 5-year period (1 January 2012 to 31 December 2016). Data obtained included the type of marketed intervention, direction of effect and declaration of both industry sponsorship and conflict of interest.ResultsEighty-four RCTs published in 23 scientific journals were included with the highest percentage in the American Journal of Dentofacial Orthopedics (AJO-DO) (23.8%), followed by the European Journal of Orthodontics (EJO) (14.3%), Journal of Orthodontics (JO) (10.7%) and Angle Orthodontist (AO) (10.7%). Overall, 45% (38/84) of clinical trials assessed involved analysis of marketed products after their introduction. Interventions to improve oral health or circumvent the risk of iatrogenic damage, such as white spot lesions, were most commonly assessed (15.8%), with the relative merits of non-surgical adjuncts (14.1%) and other orthodontic auxiliaries (13.1%) also frequently evaluated. In 44% of RCTs, a positive effect of the marketed intervention was not reported. Industry sponsorship of the research was declared in 9.5% RCTs. No significant associations between the direction of the effect and both declaration of industry sponsorship (p = 0.56) and conflict of interest (p = 0.96) were detected. Moreover, for marketed and non-marketed products, no significant associations for both declaration of industry sponsorship (p = 0.44) and conflict of interest (p = 0.28) were found.ConclusionsAlmost half of orthodontic clinical trials over the past 5 years involve analysis of marketed products after their introduction. The results highlight a potential source of waste in orthodontic research emanating from existing approaches to licensing and marketing of orthodontic products.

Iranian Nephrology and Urology Research Output in the Past Two Decades: A Bibliographic Analysis of Medline Database.

Context:We performed a bibliometric search to evaluate the number of papers published in the field of nephrology and urology by Iranian researchers in the past two decades.Evidence Acquisition:We did an online search in abstract/title part of articles with 129 keywords such as kidney, renal, hemodialysis, transplant, nephrology, glomerulonephritis, ureteral, nephrolithiasis, and etc. Endnote software version 7 was used to search articles published in PubMed database from November 1993 to November 2013. Those articles in which Iran was the affiliation of at least one of the authors were selected. These articles in the field of nephrology and urology were analyzed regarding the name of originated institution, field of study, total number of publications, type of study, collaboration rate of Iranian nephrologist and urologists for every year, annual sharing of Iranian articles in five journals with highest impact factor (IF) and journal IF.Results:The total number of publications in the field of nephrology and urology was 3,771 (average of 189 papers per year). Most of the Iranian nephrology and urology papers were from the capital city, Tehran (50.03%). There was an increasing trend in the number of publications over the years. Most papers were about transplantation (44.6%), nephrology (20.9%) and hemodialysis (16.4%). Of all, 53.7% were retrospective articles, whereas the proportion of clinical trials was relatively small (10.8%).Conclusions:Although Iranian publications in the field of nephrology and urology have had a considerable and significant increase in the recent years amongst the Middle Eastern countries, there is a wide distance to be a science exporter country.

Trends in global clinical trial registration: an analysis of numbers of registered clinical trials in different parts of the world from 2004 to 2013

ObjectivesTo analyse developments (and their causes) in the number and proportion of clinical trials that were registered in different parts of the world after the International Committee of Medical Journal...

Análisis Bibliométrico de la producción Científica Internacional sobre atención primaria

ObjectiveTo present an international view of the scientific production in the field of primary care in the period 1985-2004. DesignRetrospective, observational study. Bibliometric analysis. LocationMedline database. WebSPIRS access through version 4.3. We analysed two periods: 2000-2004 (cross-sectional) and 1985-2004 (developmental). Search based on “MeSH Major” with descriptors: Primary Care or Primary Health Care, Family Practice, Family Physicians, Nurse Practitioners, and Entry Terms associated with these. Key MeasuresWe analysed the bibliometric indicators of production, circulation, dispersion, and visibility. ResultsIn 2000-2004, published 20911 articles were published, 0.73% of total production. There was a growth rate (1985-2004) of 221%, which was 2.4 times more than average. Transience rate was 83.17%. The English language is predominant (88.81%) over 34 languages, with Spanish being the next (2.6%). Sixteen countries produce 95.67% of the articles. Spain occupies the 7th place. Universities (52%) are the most productive institutions. There are 1074 different journals with the 10 most productive being: Br J Gen Pract, Adv Nurse Pract, Aust Fam Physician, Fam Pract, Fam Med, BMJ, Aten Primary Health Serv J, Can Fam Physician, J Fam Pract. The proportion of trials (5.43%) in PC is similar to other disciplines and has increased by 453% (1985-2004). ConclusionsThe production of Primary care is about 1% of the total scientific output, with a noticeable and higher than average growth over 20 years. The proportion of clinical trials is similar to other disciplines. Although Anglo-Saxon countries and journals are the highest producers, Spain and the journal Aten Primaria is among those highlighted.

Cancer clinical trial acronyms presented at ASCO Annual Meetings.

e16558 Background: A recent review of clinical trials found that trials with acronym names were more likely to have industry involvement, enroll more patients and have less follow up time. Given the rapidly evolving landscape of cancer drugs over the last ten years, we performed a review of ASCO abstracts to identify trends in the coerciveness of trial names over time. Methods: Abstracts from the ASCO Annual Meetings from 2000-2009 were reviewed. Abstracts were included if the title contained a study name with an acronym. Details of each study including the number of abstracts, industry involvement, type of cancer, number of subjects, randomization, and type of primary endpoint, were collected. Coerciveness of the acronym was determined by assigning acronyms to not coercive (NC) or probably coercive (PC) categories based on past literature. Results: Of the 117,675 abstracts reviewed, 329 abstracts were identified with acronyms in the title. Of these, 219 studies were identified with 193 separate study names. Of these 219 trials, 14 represented the same trial and 3 shared the same name but were unrelated. 157 of the acronyms could be deemed real words, 53 could be read as possible words, and 9 were basic acronyms which could not be classified as words. 128 acronyms were considered NC and 91 were considered PC. There was no trend toward more frequent acronym use over time (approximately 0.1% each year). Industry involvement appeared to be more prevalent in PC trials with 38.5% (n = 35) trials having involvement compared to only 25.8% (n = 33) for NC trials. 20.8% (n = 19) of these PC titles first appeared from 2000-2004 compared to 79.2% (n = 72) titles appearing first from 2005- 2009. Breast cancer was the most common cancer for NC acronyms 29.7% (n = 36) while colorectal cancer was the most common cancer with a PC trial name 24.2 % (n = 22). The most common trial designs were phase II for NC trials (28.9%, n = 37), and phase III for PC trials with (35.2%, n = 32). Conclusions: This study demonstrates that although the proportion of clinical trials in oncology with acronym names has not increased over the past ten years, there has been an increase in the proportion of studies with possibly coercive titles and that those studies are more often industry sponsored. No significant financial relationships to disclose.

Do we learn the right things from clinical trials?

Starting with what is considered to be the first clinical trial – the 1944 Patulin trial [1] – clinical trials have been seen to represent a transitional state between the medicine of impressions and evidence-based medicine. Clinical trials are not research. As B. Fisher stated: “Like an electron microscope or a flow cytometer, they are mechanisms for conducting research” [2]. In other words, clinical trials are instruments to test hypotheses. They can be carried out for a number of reasons, such as to obtain approval for a new drug, to study a diagnostic tool, to establish differences between two similar diseases, to investigate a mechanism of action, to compare two drugs with a similar indication, among many others. The planning and execution of modern clinical trials are the result of a series of innovations that have been introduced into their design over a period of almost 100 years. These include the introduction of controls or comparators, a fundamental concept in a clinical trial as comparison is the rule in medicine, which always deals with relations rather than absolute results; randomization, which is important for avoiding the unconscious selection of patients in the various groups; the placebo, as an effect that may happen by chance or by suggestion; statistics, for calculating how many patients are needed in a clinical trial and for evaluating the results; strict criteria by which to include or exclude patients; monitoring, as a means of quality control; ethics to ensure that fundamental rights are not infringed. A review on clinical trials may therefore have different facets. This review focuses on clinical trials on drugs for the following reasons: (1) the largest proportion of clinical trials comprise drug trials because they represent the conduit for introducing new drugs onto the commercial market, (2) drug trials bear a great responsibility in terms of determining the best therapy for patients and (3) the results of drug trials are economically very important to a variety of stakeholders. Controlled clinical trials are at present the best way “to demonstrate a known truth, examine the validity of a hypothesis or determine the efficacy of something previously untried” [3], and they will remain the main approach for proving a relationship between cause and effect for a long time. By looking at what has been done in a less-than-perfect manner, it may be possible to acquire knowledge that will facilitate the improvement of future clinical trials.

Bayesian sample size calculations for a non-inferiority test of two proportions in clinical trials

In the process of clinical trials and health-care evaluation, Bayesian approaches have increasingly become the center of attention. In this article, sample size calculations for a non-inferiority test of two independent binomial proportions in a clinical trial are considered in a Bayesian framework. The hybrid Neyman–Pearson–Bayesian (hNPB) probability, the conditionally Bayesian (cB) probability and the unconditionally Bayesian (uB) probability are formulated through a conjugate normal analysis. The sample sizes are calculated based on formulas where normal prior distributions are assumed, and are compared with the Neyman–Pearson (NP) sample size. Our results show that the sample size based on the hNPB probability allows us to critically evaluate the appropriateness of the NP sample size. It is suggested that the sample size calculated based on the cB probability formula is smaller than the NP sample size.

Evidence-Based Medicine Is Not For Kids † 1913

Parents, payors and preceptors demand proof for what we do, but current, best evidence of efficacy for diagnostic and therapeutic manuevers is hard to come by. New reports appear daily - by the tens of thousands each year - but the busy clinician has little time, and often too few skills to evaluate them critically. A journal is now published by the American College of Physicians whose title [Evidence-Based Medicine(EBM)], editors [Haynes & Sackett], and statement of purpose [Vol 1,p 2] have promised valuable help by providing critical reviews with special focus on practical applications. We examined EBM's content for evidence of a selection bias against pediatric journals and pediatric content during its first year of publication. EBM abstracted and analyzed in detail 143 original articles from 35 peer-reviewed journals from 11/95 to 10/96. Of these, 98 [68.5%] were from 12 General-Medical [GM], 15[10.5%] from 8 Subspecialty-Medical [SM], 22 [15.4%] from 12 Other-Specialty[OS] and 8 [5.6%] from 3 Pediatric [P] journals. EBM abstracted another 7 articles on pediatric subjects,all from GM journals. Thus, < 6% of EBM abstracts were from P journals while almost half [7/15] of citations concerning pediatric topics were from GM journals. A total 17,558 original articles were published in the 35 journals reviewed by EBM as a source for their abstracts. Articles from the GM journals were three-times more likely to be abstracted and presented in detail [98/5,785 or 16.9/1,000] than those from the P journals [8/1,235 or 6.4/1,000; χ2=9.025, p=<0.005. This negative bias was equally evident for work from OS [22/5,488 or 4.0/1,000] and even from SM journals [15/5,050 or 3.0/1,000]. This bias was not explained by the proportion of clinical trials [CT], meta-analyses [MT], or practice guidelines [PG] between journals. The proportion [CT+MA+PG/total original articles] for the GM journals [738/5,785 or 12.8%] was similar to that for the OS [750/5,488 or 13.7%] and for the SM journals [700/5,050 or 13.9%]. In fact, the proportion was significantly higher for the P journals [230/1,235 or 18.6%; χ2=28.971, p<0.001. Yet, EBM abstracted articles four-times more commonly from GM [98/738 or 132.8/1,000 CT+MA+PG] than from P journals [8/230 or 34.8/1,000 CT+MA+PG; χ2=18.293, p<0.001]. Once again, this negative bias was also evident for work from the OS [22/750 or 29.3/1,000] and from the SM journals [15/700 or 21.4/1,000]. In summary, it is clear from these findings that the [unstated] purpose of EBM is to aid clinicians who engage in general medical practice. Those who practice other forms of primary care, specialty, or subspecialty medicine must look elsewhere for help.

Optimal placebo response rates for comparing two binomial proportions.

This paper considers the problem of comparing proportions in clinical trials where two parallel groups are studied. The question of what will happen to the response rate of an active therapy if some external factor affects the placebo rate is asked and two different, but similar, answers are proposed. Assuming either a logistic or probit model has substantial influence on the sample size required to detect differences and optimal comparisons are suggested. The idea is illustrated by a clinical trial of treatment for peptic ulcer and suggestions for other possible applications are presented.