Statistical Power for Analyses of Changes in Randomized Controlled Trials

Effectiveness of adding rehabilitation of cervical related sensorimotor control to manual therapy and exercise for neck pain: A randomized controlled trial

BACKGROUND

Local neck treatments and sensorimotor training can improve cervical proprioception and balance, but it remains unclear what treatments and treatment combination achieve the best outcomes.

OBJECTIVES

To investigate the most effective interventions to improve disturbances in joint position sense (JPS) and balance and their effects on neck pain, dizziness and related features in the short- and long-terms.

DESIGN

2x2 factorial, randomized controlled trial.

METHODS

Participants with neck pain (n = 152) were randomly allocated to one of four intervention groups: i) local neck treatment (NT), ii) NT + JPS/oculomotor exercises (JPS/OC), iii) NT + balance exercises, and iv) all treatments. Participants received 12 treatments over 6 weeks. Primary outcomes were postural sway and joint position error. Secondary outcomes included gait speed, dizziness, pain intensity and disability, cervical range of motion, functional ability, and quality of life. Outcome measures were taken at baseline, posttreatment and 3-, 6- and 12-month follow-ups.

RESULTS

All four interventions resulted in short- and long-term improvements in primary and secondary outcomes with medium to large effect sizes, but JPS and balance in neck torsion improved most with the addition of a combined program of JPS/OC + balance exercises to NT while balance in neck neutral improved most with the addition of balance exercises. Adding sensorimotor training was also more effective in maintaining levels of improvement in neck pain and disability at 6- and 12-months. Effect sizes of additional treatment benefits were medium to large.

CONCLUSIONS

Adding specific training of JPS/OC and balance to NT best addresses deficits in cervical proprioception and balance.

CLINICAL TRIAL REGISTRATION NUMBER

xxxxx.

Effect of Low Dose Docosahexaenoic Acid-Rich Fish Oil on Plasma Lipids and Lipoproteins in Pre-Menopausal Women: A Dose⁻Response Randomized Placebo-Controlled Trial

Omega-3 long chain polyunsaturated fatty acid (n-3 LCPUFA) supplementation has been shown to improve plasma lipid profiles in men and post-menopausal women, however, data for pre-menopausal women are lacking. The benefits of intakes less than 1 g/day have not been well studied, and dose–response data is limited. The aim of this study was to determine the effect of low doses of docosahexaenoic acid (DHA)-rich tuna oil on plasma triglyceride (TG) lowering in pre-menopausal women, and investigate if low dose DHA-rich tuna oil supplementation would increase the low-density lipoprotein (LDL) and high-density lipoprotein (HDL) particle sizes. A randomized, double-blind, placebo-controlled trial was conducted, in which 53 healthy pre-menopausal women with mildly elevated plasma TG levels consumed 0, 0.35, 0.7, or 1 g/day n-3 LCPUFA as HiDHA™ tuna oil or placebo (Sunola oil) capsules for 8 weeks. Supplementation with 1 g/day n-3 LCPUFA, but not lower doses, reduced plasma TG by 23% in pre-menopausal women. This was reflected in a dose-dependent reduction in very-low-density lipoprotein (VLDL)-TG (R² = 0.20, p = 0.003). A weak dose-dependent shift in HDL (but not LDL) particle size was identified (R² = 0.05, p = 0.04). The results of this study indicate that DHA-rich n-3 LCPUFA supplementation at a dose of 1 g/day is an effective TG-lowering agent and increases HDL particle size in pre-menopausal women.

Testing the relation between percentage change and baseline value

Testing the relation between percentage change and baseline value has been controversial, but it is not clear why this practice may yield spurious results. In this paper, we first explained why the usual testing of the relation between percentage change and baseline value is inappropriate and then demonstrated how the appropriate null hypothesis could be formulated. We also proposed a simple procedure for testing the appropriate null hypothesis based on the assumption that when there is no relation between percentage change and baseline value, the coefficients of variation for repeated measurements of a random variable should remain unchanged. Two examples were used to demonstrate how the usual testing gave rise to misleading results, whilst results from our simple test were in general consistent with those from simulations. We also undertook simulations to investigate the impact of measurement errors on the performance of the proposed test. Results suggested the type-I error rates increased with the magnitude of measurement errors, whilst the statistical power to detect a genuine relation decreased. The usual approach to testing the relation between percentage change and baseline value tended to yield misleading results and should be avoided.

Physical therapy intervention in patients with non-cardiac chest pain following a recent cardiac event: A randomized controlled trial

OBJECTIVES

To assess the effect of two different physical therapy interventions in patients with stable coronary heart disease and non-cardiac chest pain.

METHODS

A randomized controlled trial was carried out at a university hospital in Norway. A total of 30 patients with known and stable coronary heart disease and self-reported persistent chest pain reproduced by palpation of intercostal trigger points were participating in the study. The intervention was deep friction massage and heat pack versus heat pack only. The primary outcome was pain intensity after the intervention period and 3 months after the last treatment session, measured by Visual Analogue Scale, 0 to 100. Secondary outcome was health-related quality of life.

RESULTS

Treatment with deep friction massage and heat pack gave significant pain reduction compared to heat pack only (-17.6, 95% confidence interval: -30.5, -4.7; p < 0.01), and the reduction was persistent at 3 months' follow-up (-15.2, 95% confidence interval: -28.5, -1.8; p = 0.03). Health-related quality of life improved in all three domains in patients with no significant difference between groups.

CONCLUSION

Deep friction massage combined with heat pack is an efficient treatment of musculoskeletal chest pain in patients with stable coronary heart disease.

Reanalysis of morphine consumption from two randomized controlled trials of gabapentin using longitudinal statistical methods

Background

Postoperative pain management in total joint replacement surgery remains ineffective in up to 50% of patients and has an overwhelming impact in terms of patient well-being and health care burden. We present here an empirical analysis of two randomized controlled trials assessing whether addition of gabapentin to a multimodal perioperative analgesia regimen can reduce morphine consumption or improve analgesia for patients following total joint arthroplasty (the MOBILE trials).

Methods

Morphine consumption, measured for four time periods in patients undergoing total hip or total knee arthroplasty, was analyzed using a linear mixed-effects model to provide a longitudinal estimate of the treatment effect. Repeated-measures analysis of variance and generalized estimating equations were used in a sensitivity analysis to compare the robustness of the methods.

Results

There was no statistically significant difference in morphine consumption between the treatment group and a control group (mean effect size estimate 1.0, 95% confidence interval −4.7, 6.7, P=0.73). The results remained robust across different longitudinal methods.

Conclusion

The results of the current reanalysis of morphine consumption align with those of the MOBILE trials. Gabapentin did not significantly reduce morphine consumption in patients undergoing major replacement surgeries. The results remain consistent across longitudinal methods. More work in the area of postoperative pain is required to provide adequate management for this patient population.

The use of shear rate-diameter dose-response curves as an alternative to the flow-mediated dilation test

The brachial artery flow-mediated dilation test (FMD) is the non-invasive gold-standard used to test endothelial function. Reduced FMD precedes the development of atherosclerosis and provides an early marker for predicting future cardiovascular disease events. Although, this test is of high potential, it is somewhat limited by poor reproducibility. By utilizing hand warming and grip exercise combined with hierarchical linear modeling, shear rate-diameter dose-response curves may provide a novel and more accurate way to assess endothelial function in humans. Shear rate-diameter dose-response curves could potentially improve upon the traditional FMD measurement and serve as a superior clinical and research tool for assessing cardiovascular disease risk in a variety of populations. The current paper presents testable hypotheses and methodology for assessing the validity and reliability of an alternative to the current FMD test.

Empirical comparison of four baseline covariate adjustment methods in analysis of continuous outcomes in randomized controlled trials

Background

Although seemingly straightforward, the statistical comparison of a continuous variable in a randomized controlled trial that has both a pre- and posttreatment score presents an interesting challenge for trialists. We present here empirical application of four statistical methods (posttreatment scores with analysis of variance, analysis of covariance, change in scores, and percent change in scores), using data from a randomized controlled trial of postoperative pain in patients following total joint arthroplasty (the Morphine COnsumption in Joint Replacement Patients, With and Without GaBapentin Treatment, a RandomIzed ControlLEd Study [MOBILE] trials).

Methods

Analysis of covariance (ANCOVA) was used to adjust for baseline measures and to provide an unbiased estimate of the mean group difference of the 1-year postoperative knee flexion scores in knee arthroplasty patients. Robustness tests were done by comparing ANCOVA with three comparative methods: the posttreatment scores, change in scores, and percentage change from baseline.

Results

All four methods showed similar direction of effect; however, ANCOVA (−3.9; 95% confidence interval [CI]: −9.5, 1.6; P=0.15) and the posttreatment score (−4.3; 95% CI: −9.8, 1.2; P=0.12) method provided the highest precision of estimate compared with the change score (−3.0; 95% CI: −9.9, 3.8; P=0.38) and percent change (−0.019; 95% CI: −0.087, 0.050; P=0.58).

Conclusion

ANCOVA, through both simulation and empirical studies, provides the best statistical estimation for analyzing continuous outcomes requiring covariate adjustment. Our empirical findings support the use of ANCOVA as an optimal method in both design and analysis of trials with a continuous primary outcome.

Bias, precision and statistical power of analysis of covariance in the analysis of randomized trials with baseline imbalance: a simulation study

Background

Analysis of variance (ANOVA), change-score analysis (CSA) and analysis of covariance (ANCOVA) respond differently to baseline imbalance in randomized controlled trials. However, no empirical studies appear to have quantified the differential bias and precision of estimates derived from these methods of analysis, and their relative statistical power, in relation to combinations of levels of key trial characteristics. This simulation study therefore examined the relative bias, precision and statistical power of these three analyses using simulated trial data.

Methods

126 hypothetical trial scenarios were evaluated (126 000 datasets), each with continuous data simulated by using a combination of levels of: treatment effect; pretest-posttest correlation; direction and magnitude of baseline imbalance. The bias, precision and power of each method of analysis were calculated for each scenario.

Results

Compared to the unbiased estimates produced by ANCOVA, both ANOVA and CSA are subject to bias, in relation to pretest-posttest correlation and the direction of baseline imbalance. Additionally, ANOVA and CSA are less precise than ANCOVA, especially when pretest-posttest correlation ≥ 0.3. When groups are balanced at baseline, ANCOVA is at least as powerful as the other analyses. Apparently greater power of ANOVA and CSA at certain imbalances is achieved in respect of a biased treatment effect.

Conclusions

Across a range of correlations between pre- and post-treatment scores and at varying levels and direction of baseline imbalance, ANCOVA remains the optimum statistical method for the analysis of continuous outcomes in RCTs, in terms of bias, precision and statistical power.

Key statistical and analytical issues for evaluating treatment effects in periodontal research

Statistics is an indispensible tool for evaluating treatment effects in clinical research. Due to the complexities of periodontal disease progression and data collection, statistical analyses for periodontal research have been a great challenge for both clinicians and statisticians. The aim of this article is to provide an overview of several basic, but important, statistical issues related to the evaluation of treatment effects and to clarify some common statistical misconceptions. Some of these issues are general, concerning many disciplines, and some are unique to periodontal research. We first discuss several statistical concepts that have sometimes been overlooked or misunderstood by periodontal researchers. For instance, decisions about whether to use the t-test or analysis of covariance, or whether to use parametric tests such as the t-test or its non-parametric counterpart, the Mann-Whitney U-test, have perplexed many periodontal researchers. We also describe more advanced methodological issues that have sometimes been overlooked by researchers. For instance, the phenomenon of regression to the mean is a fundamental issue to be considered when evaluating treatment effects, and collinearity amongst covariates is a conundrum that must be resolved when explaining and predicting treatment effects. Quick and easy solutions to these methodological and analytical issues are not always available in the literature, and careful statistical thinking is paramount when conducting useful and meaningful research.

Can existing drugs approved for other indications retard renal function decline in patients with type 1 diabetes and nephropathy?

Mounting evidence from human, animal, and in vitro studies indicates that existing drugs, developed to treat other disorders, also might be effective in preventing or slowing the progression of diabetic nephropathy to end-stage renal disease. Examples of such drugs include the urate-lowering agent allopurinol, the anti-tumor necrosis factor agents etanercept and infliximab, and the immunomodulating drug abatacept. Because some of these medications are already on the market and have been used for a number of years for other indications, they can be tested immediately in human beings for a beneficial effect on renal function in diabetes. Special emphasis should be placed on evaluating the use of these drugs early in the course of diabetic nephropathy when renal damage is most likely to be reversible and interventions can yield the greatest delay to end-stage renal disease.

Use of ultrasound for non-invasive assessment of flow-mediated dilation

The pathological complications of atherosclerosis, namely heart attacks and strokes, remain the leading cause of mortality in the Western world. Preceding atherosclerosis is endothelial dysfunction. There is therefore interest in the application of non-invasive clinical tools to assess endothelial function. The flow-mediated dilation (FMD) test is the standard tool used to assess endothelial function. Reduced FMD is an early marker of atherosclerosis and has been noted for its capacity to predict future cardiovascular disease events. This review discusses the measurement of endothelial function using ultrasound, with a focus on the FMD technique.

An assessment of quality characteristics of randomised control trials published in dental journals

OBJECTIVES

The purpose of this study was to investigate the quality of reporting of randomised clinical trials (RCTs) published in dental specialty journals.

METHODS

The journals possessing the highest impact factor (2008 data) in the six major dental specialties were included in the study. The contents of the 24 most recent issues of each journal were hand-searched and research articles identified as randomised controlled trials (RCTs) were selected. Quality evaluation was performed using the modified Consolidated Standards of Reporting Trials (CONSORT) statement checklist. The data were analysed using descriptive statistics followed by univariate and multivariate examination of statistical associations (alpha=0.05).

RESULTS

Ninety-five RCTs were identified with generally suboptimal scores on quality reporting on key CONSORT areas. Significant differences were found among journals with the Journal of Clinical Periodontology achieving the highest score, followed by the American Journal of Orthodontics and Dentofacial Orthopedics. There was a positive association between quality score and number of authors, involvement of statistician/epidemiologist, and multicentre trials.

CONCLUSIONS

The quality scores of RCTs in major dental journals are considered suboptimal in key CONSORT areas. This receives critical importance considering that improved quality of RCTs is a fundamental prerequisite for improved dental care.

Comparison of the effects of treatment of peri-implant infection in animal and human studies: systematic review and meta-analysis

OBJECTIVE

The main objective of this systematic review is to compare the effects of treatment of peri-implant infection between animal and human studies.

MATERIAL AND METHODS

A literature search was conducted using the Medline, Cochrane Central Register of Controlled Trials, and Latin American and Caribbean Health Sciences Literature databases up to and including May 2008. In addition, bibliographies of systematic reviews on peri-implant diseases were searched manually. Non-surgical and surgical treatments of peri-implantitis/mucositis in animal models or human studies were compared. Meta-analysis was conducted to investigate the difference between the reported treatment effects in animal and human studies. Changes in probing pocket depth (PPD) and probing attachment level (PAL) from baseline measurements were used as measures of outcome. Single-level and multilevel meta-regression analysis was performed by taking into account the different follow-up times of the studies included.

RESULTS

The single-level and multilevel random-effects meta-analysis showed that the difference in PPD reduction [0.31 mm, 95% confidence interval (CI): -0.27, 0.88] and in PAL gain (0.21 mm, 95% CI: -0.47, 0.88) between animal and human studies was not statistically significant. The random-effects meta-regression suggested that studies with longer follow-up times revealed greater PPD reduction (0.25 mm per month, 95% CI: 0.14, 0.35). However, when the different follow-up times were taken into account, these differences became greater. Substantial heterogeneity between studies was found in the meta-analyses (I(2)=97.6% for animal studies and 99.9% for human studies).

CONCLUSION

There was great heterogeneity between human and animal studies in terms of study designs and treatment procedures. Therefore, the results from this meta-analysis should be interpreted with caution. Heterogeneity between studies and its causes merit further investigations.

Ultrasonographic measurements of the ulnar nerve at the elbow: role of confounders

OBJECTIVE

The purpose of this study was to identify factors confounding high-resolution ultrasonographic measurements of the ulnar nerve to test their influence when discriminating between limbs affected and unaffected by ulnar nerve entrapment (UNE) at the elbow.

METHODS

High-resolution ultrasonographic measurements of ulnar nerve dimensions at the elbow were compared between 2 groups of subjects: symptomatic and asymptomatic for UNE. Rank analysis of covariance regression tests were performed to determine whether significant differences existed between the 2 groups. The changing coefficient method (using rank analysis of covariance tests) was used to test for potential confounding effects of age, weight, height, body mass index, sex, limb sidedness, limb handedness, and nerve mobility. These tests were repeated for each measurement while controlling for the identified confounders. Exact 2-tailed Wilcoxon signed rank tests were performed to test for significant differences between measurements of the diameter of the ulnar nerve with the elbow in full extension and full flexion.

RESULTS

Age, weight, body mass index, sex, and elbow position were shown to have confounding influences on high-resolution ultrasonographic measurements of the ulnar nerve. No confounding effect was apparent for limb sidedness or dominance. Cross-sectional area and long-axis diameter measurements demonstrated significant differences between nerves with and without UNE after controlling for confounders.

CONCLUSIONS

Two cross-sectional measurements (area and maximum cross-sectional diameter) of the ulnar nerve, made at the level of the medial epicondyle, were found to be robust discriminators between nerves with and without UNE. In the absence of normative reference values of the ulnar nerve, the contralateral limb may be used as the comparative control.

Simpson's Paradox, Lord's Paradox, and Suppression Effects are the same phenomenon--the reversal paradox

This article discusses three statistical paradoxes that pervade epidemiological research: Simpson's paradox, Lord's paradox, and suppression. These paradoxes have important implications for the interpretation of evidence from observational studies. This article uses hypothetical scenarios to illustrate how the three paradoxes are different manifestations of one phenomenon – the reversal paradox – depending on whether the outcome and explanatory variables are categorical, continuous or a combination of both; this renders the issues and remedies for any one to be similar for all three. Although the three statistical paradoxes occur in different types of variables, they share the same characteristic: the association between two variables can be reversed, diminished, or enhanced when another variable is statistically controlled for. Understanding the concepts and theory behind these paradoxes provides insights into some controversial or contradictory research findings. These paradoxes show that prior knowledge and underlying causal theory play an important role in the statistical modelling of epidemiological data, where incorrect use of statistical models might produce consistent, replicable, yet erroneous results.

Comparison of Interdental Cleaning Methods: A Randomized Controlled Trial

BACKGROUND

Although interdental cleaning is an integral component of home plaque control for periodontally involved patients, limited data exist on the periodontal benefits of commonly used interdental cleaning methods before definitive root surface debridement is undertaken. Therefore, the aim of this study was to compare the effects of a customized interdental brushing technique and a customized flossing technique on clinical periodontal outcomes prior to root surface debridement in chronic periodontitis cases.

METHODS

This was a single-blind, randomized controlled clinical trial. Seventy-seven patients with chronic periodontitis were measured for plaque, relative interdental papillae level, Eastman interdental bleeding index, probing depths, and bleeding on probing at interdental sites and underwent a 10-minute hand scaling to remove easily accessible calculus deposits. Before group allocation, patients were advised on toothbrushing and instructed in two customized methods of interdental cleaning involving dental floss and precurved interdental brushes. Materials were supplied after random allocation. Participants were recalled at 6 and 12 weeks for clinical measurements, reinforcement of instructions, and fresh materials.

RESULTS

There were significant reductions from baseline for all indices in both groups (P <0.01). At 6 weeks, the interdental brush group improved more than the floss group in every parameter (P <0.05). By 12 weeks, the changes in plaque, papillae level, and probing depths were significantly greater in the interdental brush group than the floss group (P <0.01).

CONCLUSION

This trial demonstrated that patients were able to improve clinical periodontal outcomes by interdental cleaning, particularly with interdental brushes, even before thorough root surface debridement was undertaken.

Evaluating the quality of active-control trials in periodontal research

AIM

The increasing popularity of randomized-controlled trials (RCTs) has raised the issue of their quality. Frequently overlooked are the differences between superiority and equivalence trials. The purpose of this study was to apply specific methodological criteria to evaluate the quality of active-control trials using studies that compared guided tissue regeneration (GTR) with enamel matrix derivatives (EMD).

MATERIALS AND METHODS

Seven RCTs were identified in the literature. Standard methodological criteria and seven additional criteria for trials using active-control groups were used to evaluate the quality of the seven RCTs.

RESULTS

Two trials were considered as superiority trials. The remaining five provided no clear statement of their research aim. However, two claimed that EMD and GTR were equally effective, because their results failed to show a significant difference between EMD and GTR. Most trials did not meet the majority of the design criteria.

CONCLUSIONS

The general lack of compliance with quality criteria might place doubt on the value of these trials and may render any conclusions questionable. It is therefore important to distinguish clearly between superiority trials and equivalence trials, and to incorporate appropriate additional criteria in the design of future RCTs with active-control groups.

Electrical stimulation-evoked resistance exercise therapy improves arterial health after chronic spinal cord injury

STUDY DESIGN

Repeated measures training intervention.

OBJECTIVES

To evaluate the effects of neuromuscular electrical stimulation (NMES)-induced resistance exercise therapy on lower extremity arterial health in individuals with chronic, complete spinal cord injury (SCI). We define "arterial health" using three surrogate markers: (a) resting diameter, (b) flow-mediated dilation (FMD), and (c) arterial range.

SETTING

Department of Kinesiology, University of Georgia, USA.

METHODS

We assessed five 36+/-5-year-old male individuals with chronic, complete SCI before, during, and after 18 weeks of training. The quadriceps femoris muscle group of both legs were trained twice a week with 4 x 10 repetitions of unilateral, dynamic knee extensions. The health of the posterior tibial artery was assessed using a B-mode ultrasound unit equipped with a high-resolution video capture device. Proximal occlusion was used to evoke ischemia for 5 min and then for 10 min. FMD was calculated using the peak diameter change (above rest) following 5 min occlusion. Arterial range was calculated using minimum (during occlusion) and maximum diameters (post 10 min occlusion). Hierarchical linear modeling accounted for the nested (repeated measures) experimental design.

RESULTS

FMD improved from 0.08+/-0.11 mm (2.7%) to 0.18+/-0.15 mm (6.6%) (P=0.004), and arterial range improved from 0.36+/-0.28 to 0.94+/-0.40 mm (P=0.001), after 18 weeks of training. Resting diameter did not significantly change.

CONCLUSIONS

Home-based, self-administered NMES resistance exercise therapy consisting of 80 contractions/week improved FMD and arterial range. This provides evidence that resistance exercise therapy can improve arterial health after SCI, which may reduce the risk of future cardiovascular disease.

Upper vs lower extremity arterial function after spinal cord injury

BACKGROUND/OBJECTIVE

The purpose of the study was to determine whether arterial diameter, flow-mediated dilatation (FMD), and arterial range are affected by spinal cord injury (SCI). We assessed arm (radial) and leg (posterior-tibial) arteries that are comparable in size and function to determine whether (a) arterial function is reduced in individuals with SCI vs nondisabled subjects and (b) decrements to SCI arterial function are greater in the legs vs arms.

PARTICIPANTS

Eighteen men with chronic (9.8 +/- 6.3 years) SCI (T2 to T11; American Spinal Injury Association A) and 13 nondisabled subjects matched for age (33.1 +/- 4.8 vs 29.8 +/- 8.2 years old, respectively), height, and weight (BMI = 25.3 +/- 5.8 vs 26.6 +/- 5.5 kg/m2, respectively).

METHODS

Radial and posterior tibial artery B-mode ultrasound images were continuously captured to measure resting diameter, occluded diameter, and postischemic diameters. Hierarchical linear modeling accounted for the nested experimental design.

RESULTS

Individuals with SCI have lower systemic (arm + leg) FMD than nondisabled subjects (9.3% vs 12.3%, respectively; P= 0.035), primarily because of reduced leg FMD (11.5 +/- 3.1% vs 7.0 +/- 2.8% for SCI arms vs legs, respectively; P = 0.010). Persons with SCI also had lower arterial range than nondisabled subjects (0.79 vs 1.00 mm, respectively; P = 0.043), primarily because of the legs (0.81 +/- 0.09 vs 0.56 +/- 0.11 mm for SCI arms vs legs, respectively; P = 0.030).

CONCLUSION

Leg arterial function seems to deteriorate at greater rates compared to the arms for individuals with SCI. Interventions to improve cardiovascular health should include measurements taken in the legs.