Calculation of sample size for stroke trials assessing functional outcome: comparison of binary and ordinal approaches

Background Many acute stroke trials have given neutral results. Sub-optimal statistical analyses may be failing to detect efficacy. Methods which take account of the ordinal nature of functional outcome data are more efficient. We compare sample size calculations for dichotomous and ordinal outcomes for use in stroke trials. Methods Data from stroke trials studying the effects of interventions known to positively or negatively alter functional outcome – Rankin Scale and Barthel Index – were assessed. Sample size was calculated using comparisons of proportions, means, medians (according to Payne), and ordinal data (according to Whitehead). The sample sizes gained from each method were compared using Friedman 2 way ANOVA. Results Fifty-five comparisons (54 173 patients) of active vs. control treatment were assessed. Estimated sample sizes differed significantly depending on the method of calculation (Po00001). The ordering of the methods showed that the ordinal method of Whitehead and comparison of means produced significantly lower sample sizes than the other methods. The ordinal data method on average reduced sample size by 28% (inter-quartile range 14–53%) compared with the comparison of proportions; however, a 22% increase in sample size was seen with the ordinal method for trials assessing thrombolysis. The comparison of medians method of Payne gave the largest sample sizes. Conclusions Choosing an ordinal rather than binary method of analysis allows most trials to be, on average, smaller by approximately 28% for a given statistical power. Smaller trial sample sizes may help by reducing time to completion, complexity, and financial expense. However, ordinal methods may not be optimal for interventions which both improve functional outcome

Predicting long-term outcome after acute ischemic stroke: a simple index works in patients from controlled clinical trials

Background and Purpose—An early and reliable prognosis for recovery in stroke patients is important for initiation of individual treatment and for informing patients and relatives. We recently developed and validated models for predicting survival and functional independence within 3 months after acute stroke, based on age and the National Institutes of Health Stroke Scale score assessed within 6 hours after stroke. Herein we demonstrate the applicability of our models in an independent sample of patients from controlled clinical trials. Methods—The prognostic models were used to predict survival and functional recovery in 5419 patients from the Virtual International Stroke Trials Archive (VISTA). Furthermore, we tried to improve the accuracy by adapting intercepts and estimating new model parameters. Results—The original models were able to correctly classify 70.4% (survival) and 72.9% (functional recovery) of patients. Because the prediction was slightly pessimistic for patients in the controlled trials, adapting the intercept improved the accuracy to 74.8% (survival) and 74.0% (functional recovery). Novel estimation of parameters, however, yielded no relevant further improvement. Conclusions—For acute ischemic stroke patients included in controlled trials, our easy-to-apply prognostic models based on age and National Institutes of Health Stroke Scale score correctly predicted survival and functional recovery after 3 months. Furthermore, a simple adaptation helps to adjust for a different prognosis and is recommended if a large data set is available. (Stroke. 2008;39:000-000.)

Relations between zero-inflated variables in trials with horticultural crops

Certain characteristics of some vegetable crops allow multiple harvests during the production cycle; however, to our knowledge, no study has described the behavior of fruit production with progression of the production cycle in vegetable crops with multiple harvests that present data overdispersion. We aimed to characterize the data overdispersion of zero-inflated variables and identify the behavior of these variables during the production cycle of several vegetable crops with multiple harvests. Data from 11 uniformity trials were used without applying treatments; these comprise the database from the Experimental Plants Group at the Federal University of Santa Maria, Brazil. The trials were conducted using four horticultural species grown during different cultivation seasons, cultivation environments, and experimental structures. Although at each harvest, a larger number of basic units with harvest fruit was observed than units without harvest fruit, the basic unit percentage without fruit was high, generating an overdispersion within each individual harvest. The variability within each harvest was high and increased with the evolution of the production cycle of Capsicum annuum, Solanum lycopersicum var. cerasiforme, Phaseolus vulgaris, and Cucurbita pepo species. However, the correlation coefficient between the mean weight and number of harvest fruits tended to remain constant during the crop production cycle. These behaviors show that harvest management should be done individually, at each harvest, such that data overdispersion is reduced.

Can we improve the statistical analysis of stroke trials? Statistical re-analysis of functional outcomes in stroke trials

Background: Most large acute stroke trials have been neutral. Functional outcome is usually analysed using a yes or no answer, e.g. death or dependency vs. independence. We assessed which statistical approaches are most efficient in analysing outcomes from stroke trials. Methods: Individual patient data from acute, rehabilitation and stroke unit trials studying the effects of interventions which alter functional outcome were assessed. Outcomes included modified Rankin Scale, Barthel Index, and ‘3 questions’. Data were analysed using a variety of approaches which compare two treatment groups. The results for each statistical test for each trial were then compared. Results: Data from 55 datasets were obtained (47 trials, 54,173 patients). The test results differed substantially so that approaches which use the ordered nature of functional outcome data (ordinal logistic regression, t-test, robust ranks test, bootstrapping the difference in mean rank) were more efficient statistically than those which collapse the data into 2 groups (chi square) (ANOVA p<0.001). The findings were consistent across different types and sizes of trial and for the different measures of functional outcome. Conclusions: When analysing functional outcome from stroke trials, statistical tests which use the original ordered data are more efficient and more likely to yield reliable results. Suitable approaches included ordinal logistic regression, t-test, and robust ranks test.

Use of ordinal outcomes in vascular prevention trials: comparison with binary outcomes in published trials

Background and Purpose—Vascular prevention trials mostly count “yes/no” (binary) outcome events, eg, stroke/no stroke. Analysis of ordered categorical vascular events (eg, fatal stroke/nonfatal stroke/no stroke) is clinically relevant and could be more powerful statistically. Although this is not a novel idea in the statistical community, ordinal outcomes have not been applied to stroke prevention trials in the past. Methods—Summary data on stroke, myocardial infarction, combined vascular events, and bleeding were obtained by treatment group from published vascular prevention trials. Data were analyzed using 10 statistical approaches which allow comparison of 2 ordinal or binary treatment groups. The results for each statistical test for each trial were then compared using Friedman 2-way analysis of variance with multiple comparison procedures. Results—Across 85 trials (335 305 subjects) the test results differed substantially so that approaches which used the ordinal nature of stroke events (fatal/nonfatal/no stroke) were more efficient than those which combined the data to form 2 groups (P0.0001). The most efficient tests were bootstrapping the difference in mean rank, Mann–Whitney U test, and ordinal logistic regression; 4- and 5-level data were more efficient still. Similar findings were obtained for myocardial infarction, combined vascular outcomes, and bleeding. The findings were consistent across different types, designs and sizes of trial, and for the different types of intervention. Conclusions—When analyzing vascular events from prevention trials, statistical tests which use ordered categorical data are more efficient and are more likely to yield reliable results than binary tests. This approach gives additional information on treatment effects by severity of event and will allow trials to be smaller. (Stroke. 2008;39:000-000.)

Can we improve the statistical analysis of stroke trials? Statistical reanalysis of functional outcomes in stroke trials

Background and Purpose—Most large acute stroke trials have been neutral. Functional outcome is usually analyzed using a yes or no answer, eg, death or dependency versus independence. We assessed which statistical approaches are most efficient in analyzing outcomes from stroke trials. Methods—Individual patient data from acute, rehabilitation and stroke unit trials studying the effects of interventions which alter functional outcome were assessed. Outcomes included modified Rankin Scale, Barthel Index, and “3 questions”. Data were analyzed using a variety of approaches which compare 2 treatment groups. The results for each statistical test for each trial were then compared. Results—Data from 55 datasets were obtained (47 trials, 54 173 patients). The test results differed substantially so that approaches which use the ordered nature of functional outcome data (ordinal logistic regression, t test, robust ranks test, bootstrapping the difference in mean rank) were more efficient statistically than those which collapse the data into 2 groups (2; ANOVA, P0.001). The findings were consistent across different types and sizes of trial and for the different measures of functional outcome. Conclusions—When analyzing functional outcome from stroke trials, statistical tests which use the original ordered data are more efficient and more likely to yield reliable results. Suitable approaches included ordinal logistic regression, test, and robust ranks test.

Systematic reviews as a tool for planning and interpreting trials

Background Systematic reviews followed by ameta-analysis are carried out in medical research to combine the results of two or more related studies. Stroke trials have struggled to show beneficial effects and meta-analysis should be used more widely throughout the research process to either speed up the development of useful interventions, or halt more quickly research with hazardous or ineffective interventions. Summary of review. This review summarises the clinical research process and illustrates how and when systematic reviews may be used throughout the development programme. Meta-analyses should be performed after observational studies, preclinical studies in experimental stroke, and after phase I, II, and III clinical trials and phase IV clinical surveillance studies. Although meta-analyses most commonly work with summary data, they may be performed to assess relationships between variables (meta-regression) and, ideally, should utilise individual patient data. Meta-analysis techniques may alsoworkwith ordered categorical outcome data (ordinal meta-analysis) and be used to perform indirect comparisons where original trial data do not exist. Conclusion Systematic review/meta-analyses are powerful tools in medical research and should be used throughout the development of all stroke and other interventions

Chromium supplementation in patients with type 2 diabetes and high risk of type 2 diabetes: a meta-analysis of randomized controlled trials

Introduction: Chromium is an essential trace mineral for carbohydrate and lipid metabolism, which is currently prescribed to control diabetes mellitus. Results of previous systematic reviews and meta-analyses of chromium supplementation and metabolic profiles in diabetes have been inconsistent. Aim: The objective of this meta-analysis was to assess the effects on metabolic profiles and safety of chromium supplementation in type 2 diabetes mellitus and cholesterol. Methods: Literature searches in PubMed, Scopus and Web of Science were made by use of related terms-keywords and randomized clinical trials during the period of 2000-2014. Results: Thirteen trials fulfilled the inclusion criteria and were included in this systematic review. Total doses of Cr supplementation and brewer's yeast ranged from 42 to 1,000 µg/day, and duration of supplementation ranged from 30 to 120 days. The analysis indicated that there was a significant effect of chromium supplementation in diabetics on fasting plasma glucose with a weighted average effect size of -29.26 mg/dL, p = 0.01, CI 95% = -52.4 to -6.09; and on total cholesterol with a weighted average effect size of -6.7 mg/dL, p = 0.01, CI 95% = -11.88 to -1.53. Conclusions: The available evidence suggests favourable effects of chromium supplementation on glycaemic control in patients with diabetes. Chromium supplementation may additionally improve total cholesterol levels.

Family-based interventions for reducing sedentary time in youth: a systematic review of randomized controlled trials

Family involvement in interventions to reduce sedentary time may help foster appropriate long-term screen-based habits in children. This review systematically synthesized evidence from randomized controlled trials of interventions with a family component that targeted reduction of sedentary time, including TV viewing, video games and computer use, in children. MEDLINE, PubMed, PsycInfo, CINAHL and Embase were searched from inception through March 2012. Seventeen articles were considered eligible and included in the review. Studies were judged to be at low-to-moderate risk of bias. Despite inconsistent study results, level of parental involvement, rather than the setting itself, appeared an important determinant of intervention success. Studies including a parental component of medium-to-high intensity were consistently associated with statistically significant changes in sedentary behaviours. Participant age was also identified as a determinant of intervention outcomes; all three studies conducted in pre-school children demonstrated significant decreases in sedentary time. Finally, TV exposure appeared to be related to changes in energy intake rather than physical activity. Future studies should assess the effects of greater parental involvement and child age on success of sedentary behaviour interventions. More research is required to better understand the relationship between screen time and health behaviours, particularly energy intake.

Assessing randomised clinical trials of cognitive and exposure therapies for gambling disorders : a systematic review

Aims: Problem or pathological gambling is associated with significant disruption to the individual, family and community with a range of adverse outcomes, including legal, financial and mental health impairment. It occurs more frequently in younger populations, and comorbid conditions are common. Cognitive–behaviour therapy (CBT) is the most empirically established class of treatments for problematic gambling. This article reports on a systematic review and evaluation of randomised clinical trials (RCTs) concerning two core techniques of CBT: cognitive and behavioural (exposure-based) therapies. Methods: PsycINFO, MEDLINE and the Cochrane library were searched from database inception to December 2012. The CONsolidated Standards Of Reporting Trials (CONSORT) for non-pharmacological treatments was used to evaluate each study. Results: The initial search identified 104 references. After two screening phases, seven RCTs evaluating either cognitive (n = 3), exposure (n = 3) or both (n = 1) interventions remained. The studies were published between 1983 and 2003 and conducted across Australia, Canada, and Spain. On average, approximately 31% of CONSORT items were rated as ‘absent’ for each study and more than 52% rated as ‘present with some limitations’. For all studies, 70.83% of items rated as ‘absent’ were in the methods section. Conclusions: The findings from this review of randomised clinical trials involving cognitive and exposure-based treatments for gambling disorders show that the current evidence base is limited. Trials with low risk of bias are needed to be reported before recommendations are given on their effectiveness and clinicians can appraise their potential utility with confidence.

Does sleep education change sleep parameters? Comparing sleep education trials for middle school students in Australia and New Zealand

Background: Adolescents suffer daytime consequences from sleep loss. Sleep education programs have been developed in an attempt to increase sleep knowledge and/or duration. This paper presents data from three trials of the Aus-tralian Centre for Education in Sleep (ACES) program for adolescents.

Methods: The ACES program was delivered to 69 Australian adolescents in a pre-post cross-sectional design (mean age 15.2) and 29 New Zealand adolescents in a randomised control trial (mean age 14.8 years). Assessments in sleep parame-ters were undertaken at baseline and post intervention.

Results: Where sleep knowledge was evaluated (Australian trials), significant improvements were shown in all trials (All p <0.05). Where sleep duration was assessed (New Zealand trial) significant improvements were found in week and weekend sleep duration [F(1, 27)=4.26, p=0.04). Both, students and teachers found the program feasible, interesting, and educational.

Conclusions: ACES sleep education programmes can improve both sleep knowledge and sleep duration in adolescents. Improving the programme so sleep knowledge attained equates to actual sleep behaviour change are areas for future direc-tion. Collectively these findings provide encouraging signs that adolescents can improve their sleep knowledge and behav-iour with sleep education which bodes well for sleep-related health and psycho-social issues.

Statins for the treatment of depression: a meta-analysis of randomized, double-blind, placebo-controlled trials

BACKGROUND: In epidemiological studies, statins appear to benefit mood, and there are now some randomized controlled trials examining the efficacy of statins. However, the role of statins in depression remains uncertain. Thus the aim of this paper was to assess the effect of statins on depressive symptoms by performing a meta-analysis of all double-blind, randomized, placebo controlled clinical trials (RCT) conducted in subjects with depression. METHODS: A systematic search was executed using PubMed and ClinicalTrials.gov in November 30th, 2015 for all double-blind, RCT of statins versus placebo in persons with depressive symptoms. Sixty-seven potential articles were identified through search of electronic databases, of those three met inclusion criteria and were included in the meta-analysis. The outcome measure was change in Hamilton Depression Rating Scale (HDRS) scores associated with statin use. A meta-analysis was conducted and standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated. GRADE was used to assess study quality. RESULTS: The three articles included provided data on 165 participants with moderate to severe depression. Of these, 82 were randomized to statins as an adjuvant therapy to antidepressant treatment (i.e., citalopram or fluoxetine) and 83 to the placebo arm. All studies were double-blind RCTs, with a follow-up of 6-12 weeks. The statin agents evaluated were lovastatin, atorvastatin, and simvastatin. When compared to placebo, statins, as add-on to treatment as usual, largely improved depressive symptoms as assessed by the HDRS (SMD=-0.73, 95% IC -1.04 to -0.42, p<0.001, 3 between-group comparisons, n=165). No serious adverse effects were reported. CONCLUSIONS: Our results suggest that adjunctive treatment with statins could be useful for the treatment of depressive symptoms. Additional double-blind, randomised, placebo-controlled trials are necessary to settle the matter.

The science of the COAG coordinated care trials

Objectives: To explicate the organisational change agenda of the COAG coordinated care trials within the Australian health system and to illuminate the role of science in this process. Methods and Results: This article briefly outlines the COAG coordinated care trial aims and the effect of the trial as a change initiative in rural South Australia. It is proposed that although the formal trial outcomes are still not clear, the trial had significant impact upon health service delivery in some sites. The trial involved standard research methods with control and intervention groups and with key hypotheses being tested to compare the costs and service utilization profile of intervention and control groups. Formal results indicate that costs were not significantly different between intervention and control groups across all sites, but that the trial, nonetheless, had a powerful impact on the attitude and behaviours of service providers in the rural trial on Eyre Peninsula in particular. Some of the key structural changes now in place are outlined. Conclusions: The COAG trial has had many and varied impacts upon those organisations and individual providers involved with it. It is argued here that since successive initiatives had been implemented before final evaluation results were published, other agendas were served by the trial apart from those of standard scientific research and hypothesis testing. That is, the main impact of the coordinated care trial in Eyre Region at least has been change by stealth, and not through scientific research and demonstration. Implications: The COAG trials have set in train a series of structural and procedural changes in the methods of delivery and management of primary health care systems; changes that are embodied in the Enhanced Primary Care packages (EPC) and other initiatives recently introduced by the Commonwealth Government. These changes have occurred and are occurring across the system without formal evidence as to their efficacy, suggesting that other financial motives are driving these new approaches apart from the goal of improving health outcomes for consumers. Also, if science is to be used in this way to drive policy and procedural change ahead of actual outcome evidence, it is important that we examine the more subtle agendas of such research projects in future if the integrity of the scientific method is to be maintained. The occurrence of such phenomena questions the very foundation of scientific endeavour and weakens the application of scientific principles in the arena of social and political science.