Combining Statistical Evidence

Efficacy of at home monitoring of foot temperature for risk reduction of diabetes-related foot ulcer: A meta-analysis

AIMS

To perform an updated systematic review of randomised controlled trials examining the efficacy of at-home foot temperature monitoring in reducing the risk of a diabetes-related foot ulcer (DFU).

METHODS

Systematic review performed according to Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines. Risk-of-bias was assessed using version 2 of the Cochrane risk-of-bias tool. Meta-analyses were performed using random effect models. Leave-one-out sensitivity analyses and a sub-analysis excluding trials considered at high risk-of-bias assessed the consistency of the findings. The certainty of the evidence was assessed with GRADE.

RESULTS

Five randomised controlled trials involving 772 participants meeting the International Working Group on the Diabetic Foot (IWGDF) risk category 2 or 3 were included. All trials reported instructing participants to measure skin temperature at-home at six or more sites on each foot using a hand-held infra-red thermometer at least daily and reduce ambulatory activity in response to hotspots (temperature differences >2.2°C on two consecutive days between similar locations in both feet). One, one, and three trials were considered at low, moderate and high risk-of-bias, respectively. Participants allocated to at-home foot temperature monitoring had a reduced risk of developing a DFU (relative risk 0.51, 95% CI 0.31-0.84) compared to controls. Sensitivity and sub-analyses suggested that the significance of this finding was consistent. The GRADE assessment suggested a low degree of certainty in the finding.

CONCLUSIONS

At-home daily foot temperature monitoring and reduction of ambulatory activity in response to hotspots reduce the risk of a DFU in moderate or high risk people with a low level of certainty.

A Systematic Review and Meta-Analysis of the Effect of Pentagalloyl Glucose Administration on Aortic Expansion in Animal Models

Background: The aim of this systematic review was to pool evidence from studies testing if pentagalloyl glucose (PGG) limited aortic expansion in animal models of abdominal aortic aneurysm (AAA). Methods: The review was conducted according to the PRISMA guidelines and registered with PROSPERO. The primary outcome was aortic expansion assessed by direct measurement. Secondary outcomes included aortic expansion measured by ultrasound and aortic diameter at study completion. Sub analyses examined the effect of PGG delivery in specific forms (nanoparticles, periadventitial or intraluminal), and at different times (from the start of AAA induction or when AAA was established), and tested in different animals (pigs, rats and mice) and AAA models (calcium chloride, periadventitial, intraluminal elastase or angiotensin II). Meta-analyses were performed using Mantel-Haenszel’s methods with random effect models and reported as mean difference (MD) and 95% confidence intervals (CIs). Risk of bias was assessed with a customized tool. Results: Eleven studies reported in eight publications involving 214 animals were included. PGG significantly reduced aortic expansion measured by direct observation (MD: −66.35%; 95% CI: −108.44, −24.27; p = 0.002) but not ultrasound (MD: −32.91%; 95% CI: −75.16, 9.33; p = 0.127). PGG delivered intravenously within nanoparticles significantly reduced aortic expansion, measured by both direct observation (MD: −116.41%; 95% CI: −132.20, −100.62; p < 0.001) and ultrasound (MD: −98.40%; 95% CI: −113.99, −82.81; p < 0.001). In studies measuring aortic expansion by direct observation, PGG administered topically to the adventitia of the aorta (MD: −28.41%; 95% CI −46.57, −10.25; p = 0.002), studied in rats (MD: −56.61%; 95% CI: −101.76, −11.46; p = 0.014), within the calcium chloride model (MD: −56.61%; 95% CI: −101.76, −11.46; p = 0.014) and tested in established AAAs (MD: −90.36; 95% CI: −135.82, −44.89; p < 0.001), significantly reduced aortic expansion. The findings of other analyses were not significant. The risk of bias of all studies was high. Conclusion: There is inconsistent low-quality evidence that PGG inhibits aortic expansion in animal models.

Meta-analysis of the association between angiotensin pathway inhibitors and COVID-19 severity and mortality

BACKGROUND

Conflicting findings and the analysis of unpublished and retracted data have led to controversy on the safety of angiotensin-converting enzyme inhibitors and angiotensin receptor blockers in people with COVID-19 infection. This meta-analysis examined the association of prescription of angiotensin-converting enzyme inhibitors (ACEI) and angiotensin receptor blockers (ARB) with the outcome from COVID-19.

METHODS

A systematic search was conducted to find published studies that reported the outcome of COVID-19 in relation to prescription of ACEI or ARB. Two authors (MF and AD) independently screened and extracted data and assessed study quality and strength of association using standardised tools. The endpoints for the meta-analyses were severe or critical disease outcome and mortality based on standardised criteria.

RESULTS

Twenty-six studies including 8389 people prescribed ACEI or ARB and 20,989 people not prescribed these medications were included. The quality of studies varied, and the overall strength of association was poor with a high risk of confounding bias. Patients prescribed ACEI or ARB had a greater prevalence of risk factors. Meta-analysis found an association between prescription of ACEI or ARB with severe or critical disease outcome (risk ratio, RR, 1.23, 95% confidence interval, CI, 1.06 to 1.42, p = 0.006, I2 = 88%) but this association was lost in sensitivity analyses. There was no association between ACEI or ARB prescription and mortality (RR 1.18, 95% CI 0.92 to 1.50, p = 0.19, I2 = 82%).

CONCLUSIONS

This meta-analysis suggests that people prescribed ACEI or ARB more commonly had severe or critical disease outcome, but not mortality, in published cohorts of patients diagnosed with COVID-19. This finding is most likely due to a greater prevalence of risk factors in these patients rather than due to exposure to angiotensin pathway inhibitors.

Meta-analyses of randomized controlled trials reporting the effect of home foot temperature monitoring, patient education or offloading footwear on the incidence of diabetes-related foot ulcers

AIM

The aim of this study was to perform an up-to-date systematic review and meta-analysis of randomized controlled trials (RCTs) examining the efficacy of home foot temperature monitoring, patient education and offloading footwear in reducing the incidence of diabetes-related foot ulcers.

METHODS

A literature search was performed using MEDLINE, PubMed, CINAHL, Scopus and Cochrane databases to identify relevant original studies. Meta-analyses were performed using intention-to-treat principals for worst (main analysis) and best (sub-analysis) case scenarios. Leave-one-out sensitivity analyses were used to assess the consistency of findings.

RESULTS

Of 7575 unique records, 17 RCTs involving 2729 participants were included. Four tested home foot temperature monitoring (n = 468), six examined patient education (n = 823) and seven assessed offloading footwear (n = 1438). Participants' who performed home foot temperature monitoring [odds ratio (OR) 0.51, 95% confidence interval (CI) 0.31 to 0.84; n = 468] and those provided offloading footwear (OR 0.48, 95% CI 0.29 to 0.80; n = 1438) were less likely to develop a diabetes-related foot ulcer. Patient education programmes did not significantly reduce diabetes-related foot ulcer incidence (OR 0.59, 95% CI 0.29 to 1.20; n = 823). Sensitivity analyses suggested that offloading footwear findings were consistent, but home foot temperature findings were dependent on the individual inclusion of one trial. All RCTs had either high or unclear risk of bias.

CONCLUSION

This meta-analysis suggests that offloading footwear is effective in reducing the incidence of diabetes-related foot ulcers. Home foot temperature monitoring also appears beneficial but larger trials are needed (PROSPERO registration no.: CRD42019135226).

Estimation in meta-analyses of mean difference and standardized mean difference

Methods for random‐effects meta‐analysis require an estimate of the between‐study variance, τ². The performance of estimators of τ² (measured by bias and coverage) affects their usefulness in assessing heterogeneity of study‐level effects and also the performance of related estimators of the overall effect. However, as we show, the performance of the methods varies widely among effect measures. For the effect measures mean difference (MD) and standardized MD (SMD), we use improved effect‐measure‐specific approximations to the expected value of Q for both MD and SMD to introduce two new methods of point estimation of τ² for MD (Welch‐type and corrected DerSimonian‐Laird) and one WT interval method. We also introduce one point estimator and one interval estimator for τ² in SMD. Extensive simulations compare our methods with four point estimators of τ² (the popular methods of DerSimonian‐Laird, restricted maximum likelihood, and Mandel and Paule, and the less‐familiar method of Jackson) and four interval estimators for τ² (profile likelihood, Q‐profile, Biggerstaff and Jackson, and Jackson). We also study related point and interval estimators of the overall effect, including an estimator whose weights use only study‐level sample sizes. We provide measure‐specific recommendations from our comprehensive simulation study and discuss an example.

Overdispersion models for correlated multinomial data: Applications to blinding assessment

Overdispersion models have been extensively studied for correlated normal and binomial data but much less so for correlated multinomial data. In this work, we describe a multinomial overdispersion model that leads to the specification of the first two moments of the outcome and allows the estimation of the global parameters using generalized estimating equations (GEE). We introduce a Global Blinding Index as a target parameter and illustrate the application of the GEE method to its estimation from (1) a clinical trial with clustering by practitioner and (2) a meta-analysis on psychiatric disorders. We examine the impact of a small number of clusters, high variability in cluster sizes, and the magnitude of the intraclass correlation on the performance of the GEE estimators of the Global Blinding Index using the data simulated from different models. We compare these estimators with the inverse-variance weighted estimators and a maximum-likelihood estimator, derived under the Dirichlet-multinomial model. Our results indicate that the performance of the GEE estimators was satisfactory even in situations with a small number of clusters, whereas the inverse-variance weighted estimators performed poorly, especially for larger values of the intraclass correlation coefficient. Our findings and illustrations may be instrumental for practitioners who analyze clustered multinomial data from clinical trials and/or meta-analysis.

Meta-analysis of binary outcomes via generalized linear mixed models: a simulation study

Background

Systematic reviews and meta-analyses of binary outcomes are widespread in all areas of application. The odds ratio, in particular, is by far the most popular effect measure. However, the standard meta-analysis of odds ratios using a random-effects model has a number of potential problems. An attractive alternative approach for the meta-analysis of binary outcomes uses a class of generalized linear mixed models (GLMMs). GLMMs are believed to overcome the problems of the standard random-effects model because they use a correct binomial-normal likelihood. However, this belief is based on theoretical considerations, and no sufficient simulations have assessed the performance of GLMMs in meta-analysis. This gap may be due to the computational complexity of these models and the resulting considerable time requirements.

Methods

The present study is the first to provide extensive simulations on the performance of four GLMM methods (models with fixed and random study effects and two conditional methods) for meta-analysis of odds ratios in comparison to the standard random effects model.

Results

In our simulations, the hypergeometric-normal model provided less biased estimation of the heterogeneity variance than the standard random-effects meta-analysis using the restricted maximum likelihood (REML) estimation when the data were sparse, but the REML method performed similarly for the point estimation of the odds ratio, and better for the interval estimation.

Conclusions

It is difficult to recommend the use of GLMMs in the practice of meta-analysis. The problem of finding uniformly good methods of the meta-analysis for binary outcomes is still open.

Electronic supplementary material

The online version of this article (10.1186/s12874-018-0531-9) contains supplementary material, which is available to authorized users.

Beta-binomial model for meta-analysis of odds ratios

In meta-analysis of odds ratios (ORs), heterogeneity between the studies is usually modelled via the additive random effects model (REM). An alternative, multiplicative REM for ORs uses overdispersion. The multiplicative factor in this overdispersion model (ODM) can be interpreted as an intra-class correlation (ICC) parameter. This model naturally arises when the probabilities of an event in one or both arms of a comparative study are themselves beta-distributed, resulting in beta-binomial distributions. We propose two new estimators of the ICC for meta-analysis in this setting. One is based on the inverted Breslow-Day test, and the other on the improved gamma approximation by Kulinskaya and Dollinger (2015, p. 26) to the distribution of Cochran's Q. The performance of these and several other estimators of ICC on bias and coverage is studied by simulation. Additionally, the Mantel-Haenszel approach to estimation of ORs is extended to the beta-binomial model, and we study performance of various ICC estimators when used in the Mantel-Haenszel or the inverse-variance method to combine ORs in meta-analysis. The results of the simulations show that the improved gamma-based estimator of ICC is superior for small sample sizes, and the Breslow-Day-based estimator is the best for n⩾100. The Mantel-Haenszel-based estimator of OR is very biased and is not recommended. The inverse-variance approach is also somewhat biased for ORs≠1, but this bias is not very large in practical settings. Developed methods and R programs, provided in the Web Appendix, make the beta-binomial model a feasible alternative to the standard REM for meta-analysis of ORs. © 2017 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.

Inference for binomial probability based on dependent Bernoulli random variables with applications to meta-analysis and group level studies

We study bias arising as a result of nonlinear transformations of random variables in random or mixed effects models and its effect on inference in group‐level studies or in meta‐analysis. The findings are illustrated on the example of overdispersed binomial distributions, where we demonstrate considerable biases arising from standard log‐odds and arcsine transformations of the estimated probability p^, both for single‐group studies and in combining results from several groups or studies in meta‐analysis. Our simulations confirm that these biases are linear in ρ, for small values of ρ, the intracluster correlation coefficient. These biases do not depend on the sample sizes or the number of studies K in a meta‐analysis and result in abysmal coverage of the combined effect for large K. We also propose bias‐correction for the arcsine transformation. Our simulations demonstrate that this bias‐correction works well for small values of the intraclass correlation. The methods are applied to two examples of meta‐analyses of prevalence.

Which Psychological Factors are Related to HIV Testing? A Quantitative Systematic Review of Global Studies

Deciding to test for HIV is necessary for receiving HIV treatment and care among those who are HIV-positive. This article presents a systematic review of quantitative studies on relationships between psychological (cognitive and affective) variables and HIV testing. Sixty two studies were included (fifty six cross sectional). Most measured lifetime testing. HIV knowledge, risk perception and stigma were the most commonly measured psychological variables. Meta-analysis was carried out on the relationships between HIV knowledge and testing, and HIV risk perception and testing. Both relationships were positive and significant, representing small effects (HIV knowledge, d = 0.22, 95 % CI 0.14-0.31, p < 0.001; HIV risk perception, OR 1.47, 95 % CI 1.26-1.67, p < 0.001). Other variables with a majority of studies showing a relationship with HIV testing included: perceived testing benefits, testing fear, perceived behavioural control/self-efficacy, knowledge of testing sites, prejudiced attitudes towards people living with HIV, and knowing someone with HIV. Research and practice implications are outlined.