Exploring consequences of simulation design for apparent performance of methods of meta-analysis

Contemporary statistical publications rely on simulation to evaluate performance of new methods and compare them with established methods. In the context of random-effects meta-analysis of log-odds-ratios, we investigate how choices in generating data affect such conclusions. The choices we study include the overall log-odds-ratio, the distribution of probabilities in the control arm, and the distribution of study-level sample sizes. We retain the customary normal distribution of study-level effects. To examine the impact of the components of simulations, we assess the performance of the best available inverse–variance–weighted two-stage method, a two-stage method with constant sample-size-based weights, and two generalized linear mixed models. The results show no important differences between fixed and random sample sizes. In contrast, we found differences among data-generation models in estimation of heterogeneity variance and overall log-odds-ratio. This sensitivity to design poses challenges for use of simulation in choosing methods of meta-analysis.

Download Full-text

Revascularisation in older adult patients with non-ST-segment elevation acute coronary syndrome: effect and impact on 6-month mortality

European Heart Journal Acute Cardiovascular Care ◽

10.1177/2048872619849922 ◽

2019 ◽

Vol 9 (4) ◽

pp. 358-366 ◽

Cited By ~ 2

Author(s):

Alfredo Bardaji ◽

José A Barrabés ◽

Aida Ribera ◽

Héctor Bueno ◽

Antonio Fernández-Ortiz ◽

...

Keyword(s):

Acute Coronary Syndrome ◽

Confidence Interval ◽

Older Adult ◽

Odds Ratio ◽

Meta Analysis ◽

Two Stage ◽

St Segment Elevation ◽

Coronary Syndrome ◽

St Segment ◽

The Impact

Although revascularisation in non-ST-segment elevation acute coronary syndrome (NSTEACS) is associated with better outcomes, its impact in older adult patients is unclear. This is a retrospective analyses of three national NSTEACS registries conducted during the past decade in Spain. Patients aged 75 years and older were included: DESCARTES (DES; year 2002; n=534), MASCARA (MAS; 2005; n=1736) and DIOCLES (DIO; 2012; n=593). The adjusted association between revascularisation and total (inhospital and 6-month) mortality was estimated by two-stage meta-analysis (pooled effect across the three registries with inverse-variability weights) and one-stage meta-analysis (multilevel model with random effects across studies). The impact of revascularisation was assessed comparing the observed and the expected mortality based on a logistic regression model in the pooled database. Although revascularisation was associated with a lower risk of mortality in meta-analyses (two-stage: odds ratio 0.44, 95% confidence interval 0.29–0.67; one-stage: odds ratio 0.54, 95% confidence interval 0.36–0.81) and the revascularisation rate increased steadily from 2002 (DES 14.2%) to 2012 (DIO 43.7%), its impact was not patent across registries, probably because this increase was concentrated in low and medium-risk GRACE strata (tertile 1, 2 and 3: MAS 59%, 20% and 6%; DIO 64%, 39% and 19%, respectively). In conclusion, a consistent increase of revascularisation in NSTEACS in older adults was not followed by a decrease in mortality at 6 months, probably because the impact of this strategy is limited to the higher risk population, the stratum with the lowest revascularisation rate in real life.

Download Full-text

Simulation results on the performance of statistical methods in cumulative meta analysis

10.31222/osf.io/8t4pf ◽

2021 ◽

Author(s):

Elena Kulinskaya ◽

Eung Yaw Mah

Keyword(s):

Meta Analysis ◽

Temporal Trends ◽

Two Stage ◽

Type 1 Error ◽

Effect Measure ◽

Inverse Variance ◽

Study Results ◽

Heterogeneity Variance ◽

Stage 1 ◽

Weighted Estimation

Cumulative meta-analysis (CMA) is a process of updating the results of existing meta-analysis to incorporate new study results. This is a popular way to present time-varying evidence. We investigate the properties of CMA, suggest possible improvements and provide the first in-depth simulation study of the use of CMA and CUSUM methods for detection of temporal trends in random-effects meta-analysis. We use the standardized mean difference (SMD) as an effect measure of interest. For CMA, we compare the standard inverse-variance-weighted estimation of the overall effect using REML-estimated between-study variance $\tau^2$ with the sample-size-weighted estimation of the effect combined with Kulinskaya-Dollinger-Bjørkestøl (2011) (KDB) estimation of $\tau^2$. For all methods, we consider type 1 error under no shift and power under shift in the mean. To ameliorate the lack of power in CMA, we introduce the two-stage CMA, where the heterogeneity variance $\tau^2$ is estimated at stage 1 (first 5-10 studies), and the further CMA monitors a target value of effect, keeping the $\tau^2$ value fixed. We recommend the use of this two-stage CMA combined with cumulative testing for positive shift in $\tau^2$.

Download Full-text

Impact of patient and public involvement on enrolment and retention in clinical trials: systematic review and meta-analysis

BMJ ◽

10.1136/bmj.k4738 ◽

2018 ◽

pp. k4738 ◽

Cited By ~ 67

Author(s):

Joanna C Crocker ◽

Ignacio Ricci-Cabello ◽

Adwoa Parker ◽

Jennifer A Hirst ◽

Alan Chant ◽

...

Keyword(s):

Systematic Review ◽

Clinical Trial ◽

Clinical Trials ◽

Lived Experience ◽

Odds Ratio ◽

Public Involvement ◽

Meta Analysis ◽

Patient And Public Involvement ◽

Retention Rates ◽

The Impact

AbstractObjectiveTo investigate the impact of patient and public involvement (PPI) on rates of enrolment and retention in clinical trials and explore how this varies with the context and nature of PPI.DesignSystematic review and meta-analysis.Data sourcesTen electronic databases, including Medline, INVOLVE Evidence Library, and clinical trial registries.Eligibility criteriaExperimental and observational studies quantitatively evaluating the impact of a PPI intervention, compared with no intervention or non-PPI intervention(s), on participant enrolment and/or retention rates in a clinical trial or trials. PPI interventions could include additional non-PPI components inseparable from the PPI (for example, other stakeholder involvement).Data extraction and analysisTwo independent reviewers extracted data on enrolment and retention rates, as well as on the context and characteristics of PPI intervention, and assessed risk of bias. Random effects meta-analyses were used to determine the average effect of PPI interventions on enrolment and retention in clinical trials: main analysis including randomised studies only, secondary analysis adding non-randomised studies, and several exploratory subgroup and sensitivity analyses.Results26 studies were included in the review; 19 were eligible for enrolment meta-analysis and five for retention meta-analysis. Various PPI interventions were identified with different degrees of involvement, different numbers and types of people involved, and input at different stages of the trial process. On average, PPI interventions modestly but significantly increased the odds of participant enrolment in the main analysis (odds ratio 1.16, 95% confidence interval and prediction interval 1.01 to 1.34). Non-PPI components of interventions may have contributed to this effect. In exploratory subgroup analyses, the involvement of people with lived experience of the condition under study was significantly associated with improved enrolment (odds ratio 3.14v1.07; P=0.02). The findings for retention were inconclusive owing to the paucity of eligible studies (odds ratio 1.16, 95% confidence interval 0.33 to 4.14), for main analysis).ConclusionsThese findings add weight to the case for PPI in clinical trials by indicating that it is likely to improve enrolment of participants, especially if it includes people with lived experience of the health condition under study. Further research is needed to assess which types of PPI work best in particular contexts, the cost effectiveness of PPI, the impact of PPI at earlier stages of trial design, and the impact of PPI interventions specifically targeting retention.Systematic review registrationPROSPERO CRD42016043808.

Download Full-text

Meta-Analysis of the Relationship Between Ethnicity, Obesity, and Type 2 Diabetes of Adults in Urban Populations of Central America

International Journal of Public Health Science (IJPHS) ◽

10.11591/.v5i3.4796 ◽

2016 ◽

Vol 5 (3) ◽

pp. 274

Author(s):

William G Wuenstel ◽

James A. Johnson ◽

James Humphries ◽

Cheryl Samuel

Keyword(s):

Type 2 Diabetes ◽

Relative Risk ◽

Odds Ratio ◽

Meta Analysis ◽

Central American ◽

Significance Level ◽

Study Participants ◽

The Impact ◽

The Relationship

<table width="593" border="1" cellspacing="0" cellpadding="0"><tbody><tr><td rowspan="2" valign="top" width="387">The purpose of this meta-analysis was to examine the impact of ethnicity and obesity as it relates to Type-2 Diabetes (T2D) in specific Central American countries. A meta-analysis was conducted to determine the association of ethnicity, obesity, and T2D. Four studies that qualified for inclusion were identified by searching MEDLINE and PubMed databases. The studies on the association of ethnicity and T2D had a combined population resulted in 265,858 study participants. Two studies on the association of obesity and T2D had 197,899 participants. An analysis of the data was conducted utilizing the relative risk ration, odds ratio, and forest plots. The comparison of the relative risk of T2D across ethnic categories by studies range for Blacks was 1.59 to 2.74, Asians was 1.43 to 2.08, and Hispanics .92 to 2.91. The ethnic difference in the prevalence of diabetes was almost two-fold higher in all ethnic groups than among the Caucasians with a significance level of 95%. A comparison of relative risk of T2D across weight categories was significantly higher among those with a diagnosed of diabetes in all reported areas. The odds ratio was very close to the risk ratio in both ethnicity and obesity to the development of T2D. The meta-analysis findings documented that an association does exist between ethnicity and obesity to the development of type 2 diabetes.</td><td width="0" height="85"> </td></tr><tr><td width="0" height="82"> </td></tr></tbody></table>

Download Full-text

Association Between Atrial Fibrillation and Cognitive Impairment in Individuals With Prior Stroke

Stroke ◽

10.1161/strokeaha.119.027815 ◽

2020 ◽

Vol 51 (6) ◽

pp. 1662-1666 ◽

Cited By ~ 1

Author(s):

Damianos G. Kokkinidis ◽

Nikos Zareifopoulos ◽

Christina A. Theochari ◽

Angelos Arfaras-Melainis ◽

Christos A. Papanastasiou ◽

...

Keyword(s):

Atrial Fibrillation ◽

Cognitive Impairment ◽

Odds Ratio ◽

Meta Analysis ◽

Adjusted Analysis ◽

End Point ◽

History Of ◽

The Impact ◽

Prior Stroke ◽

Unadjusted Analysis

Background and Purpose— Atrial fibrillation (AF) is the most common chronic arrhythmia. Dementia and cognitive impairment (CI) are major burdens to public health. The prevalence of all 3 entities is projected to increase due to population aging. Previous reports have linked AF with a higher risk of CI and dementia in patients without prior stroke. Stroke is known to increase the risk for dementia and CI. It is unclear if AF in patients with history of stroke can further increase the risk for dementia or CI. Our purpose was to evaluate the impact of AF on risk for dementia or CI among patients with history of stroke. Methods— Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines were followed. Pubmed, Scopus, and Cochrane central were searched. The outcomes of interest were dementia, CI, and the composite end point of dementia or CI. A random-effect model meta-analysis was performed. Meta-regression analysis was also performed. Publication bias was assessed with the Egger test and with funnel plots. Results— Fourteen studies and 14 360 patients (1363 with AF) were included in the meta-analysis. In the meta-analysis of adjusted odds ratio, AF was associated with increased risk of CI (odds ratio, 1.60 [95% CI, 1.20–2.14]), dementia (odds ratio, 3.11 [95% CI, 2.05–4.73]), and the composite end point of CI or dementia (odds ratio, 2.26 [95% CI, 1.61–3.19]). The heterogeneity for the composite end point of dementia or CI was moderate (adjusted analysis). The heterogeneity for the analysis of the end point of CI only was substantial in the unadjusted analysis and moderate in the adjusted analysis. The heterogeneity for the end point of dementia only was moderate in the unadjusted analysis and zero in the adjusted analysis. Conclusions— Our results indicate that an association between AF and CI or dementia is patients with prior strokes is possible given the persistent positive associations we noticed in the unadjusted and adjusted analyses. The heterogeneity levels limit the certainty of our findings.

Download Full-text

Hepatitis B reactivation in patients with solid tumors: A systematic review and meta-analysis.

Journal of Clinical Oncology ◽

10.1200/jco.2016.34.7_suppl.138 ◽

2016 ◽

Vol 34 (7_suppl) ◽

pp. 138-138 ◽

Cited By ~ 2

Author(s):

Lisa K. Hicks ◽

Jordan J. Feld ◽

Ronak Saluja ◽

Judy Truong ◽

Adam E. Haynes ◽

...

Keyword(s):

Systematic Review ◽

Hepatitis B ◽

Solid Tumors ◽

Solid Tumor ◽

Odds Ratio ◽

Meta Analysis ◽

Hbv Reactivation ◽

Antiviral Prophylaxis ◽

Chronic Hbv ◽

The Impact

138 Background: Hepatitis B virus (HBV) affects over 250 million people worldwide. Most people with chronic HBV (HBsAg positive) have no signs or symptoms of infection. However, when exposed to immunosuppression they are at risk of HBV reactivation which can cause hepatitis, liver failure and death. The risk of HBV reactivation in patients receiving chemotherapy for solid tumors, the efficacy of antiviral prophylaxis, and the clinical impact of HBV reactivation in this setting are uncertain. Primary Aim: To estimate the risk of clinical HBV reactivation (increased HBV DNA + transaminitis) among HBsAg-positive patients administered chemotherapy for a solid tumor. Secondary Aims: To estimate the efficacy of anti-viral prophylaxis and the risk of death from HBV reactivation in patients receiving chemotherapy for solid tumors. Methods: A systematic review and meta-analysis of the English language literature on HBV reactivation was completed (OVID Medline, 1946 to Aug 2013). All citations were reviewed by two or more authors. Data from patients with hematologic malignancies were excluded. Pooled probabilities of HBV reactivation risk, death from HBV reactivation, and odds ratio for the impact of anti-viral prophylaxis were estimated with a random effects model. Results: 2,667 citations were identified; 19 were eligible for inclusion. The pooled estimate for clinical HBV reactivation in HBsAg-positive patients receiving chemotherapy for a solid tumor was 21.9% (95% CI; 16.5% to 27.3%) in those not receiving anti-viral prophylaxis, and 2.4% (95% CI 0.7% to 4.2%) in those receiving anti-viral prophylaxis. The odds ratio for clinical HBV reactivation with antiviral prophylaxis compared to no prophylaxis was 0.12 (95% CI 0.06 to 0.25). In the absence of viral prophylaxis, the risk of dying from HBV reactivation in HBsAg-positive solid tumor patients was estimated at 1.3% with a 95% CI of 0.3% to 2.3%. Conclusions: Patients with chronic HBV who are administered chemotherapy for a solid tumor appear to be at substantial risk of clinical HBV reactivation; this risk may be mitigated by anti-viral prophylaxis. In the absence of anti-viral therapy, patients may experience a small but important risk of dying from HBV reactivation.

Download Full-text

Peer influence in adolescent drinking behavior: A meta-analysis of stochastic actor-based modeling studies

PLoS ONE ◽

10.1371/journal.pone.0250169 ◽

2021 ◽

Vol 16 (4) ◽

pp. e0250169

Author(s):

Valeria Ivaniushina ◽

Vera Titkova

Keyword(s):

Peer Influence ◽

Odds Ratio ◽

Meta Analysis ◽

Drinking Behavior ◽

Selection Effect ◽

Parameter Estimates ◽

Effective Prevention ◽

Peer Selection ◽

Influence Effect ◽

Log Odds

Objectives To measure the effects of peer influence and peer selection on drinking behavior in adolescence through a rigorous statistical approach designed to unravel these interrelated processes. Methods We conducted systematic searches of electronic databases, thesis collections and conference proceedings to identify studies that used longitudinal network design and stochastic actor-oriented modeling to analyze drinking behavior in adolescents. Parameter estimates collected from individual studies were analyzed using multilevel random-effects models. Results We identified 26 articles eligible for meta-analysis. Meta-analyses for different specifications of the peer influence effect were conducted separately. The peer influence effect was positive for every specification: for average similarity (avSim) mean log odds ratio was 1.27 with 95% confidence interval [0.04; 2.49]; for total similarity (totSim) 0.46 (95% CI = [0.44; 0.48]), and for average alter (avAlt) 0.70 (95% CI = [-0.01; 1.41]). The peer selection effect (simX) was also positive: 0.46 (95% CI = [0.28; 0.63]). Conversion log odds ratio values to Cohen’s d gives estimates from 0.25 to 0.70, which is considered as medium to large effect. Conclusions Advances in methodology for social network analysis have made it possible to accurately estimate peer influence effects free from peer selection effects. More research is necessary to clarify the roles of age, gender, and individual susceptibility on the changing behavior of adolescents under the influence of their peers. Understanding the effects of peer influence should inform practitioners and policy makers to design and deliver more effective prevention programs.

Download Full-text

Ethnicity, Obesity, and Type 2 Diabetes of Adults in Urban Populations of Central America

International Journal of Public Health Science (IJPHS) ◽

10.11591/ijphs.v6i1.6529 ◽

2017 ◽

Vol 6 (1) ◽

pp. 27

Author(s):

William G Wuenstel ◽

James A. Johnson ◽

James Humphries ◽

Cheryl Samuel

Keyword(s):

Type 2 Diabetes ◽

Relative Risk ◽

Odds Ratio ◽

Ethnic Difference ◽

Meta Analysis ◽

Central American ◽

Significance Level ◽

Study Participants ◽

The Impact

The purpose of this meta-analysis was to examine the impact of ethnicity and obesity as it relates to Type-2 Diabetes (T2D) in specific Central American countries. A meta-analysis was conducted to determine the association of ethnicity, obesity, and T2D. Four studies that qualified for inclusion were identified by searching MEDLINE and PubMed databases. The studies on the association of ethnicity and T2D had a combined population resulted in 265,858 study participants. Two studies on the association of obesity and T2D had 197,899 participants. An analysis of the data was conducted utilizing the relative risk ration, odds ratio, and forest plots. The comparison of the relative risk of T2D across ethnic categories by studies range for Blacks was 1.59 to 2.74, Asians was 1.43 to 2.08, and Hispanics .92 to 2.91. The ethnic difference in the prevalence of diabetes was almost two-fold higher in all ethnic groups than among the Caucasians with a significance level of 95%. A comparison of relative risk of T2D across weight categories was significantly higher among those with a diagnosed of diabetes in all reported areas. The odds ratio was very close to the risk ratio in both ethnicity and obesity to the development of T2D.The meta-analysis findings documented that an association does exist between ethnicity and obesity to the development of type 2 diabetes.

Download Full-text

Ethnicity, Obesity, and Type 2 Diabetes of Adults in Urban Populations of Central America

International Journal of Public Health Science (IJPHS) ◽

10.11591/.v6i1.6529 ◽

2017 ◽

Vol 6 (1) ◽

pp. 28

Author(s):

William G Wuenstel ◽

James A. Johnson ◽

James Humphries ◽

Cheryl Samuel

Keyword(s):

Type 2 Diabetes ◽

Relative Risk ◽

Odds Ratio ◽

Ethnic Difference ◽

Meta Analysis ◽

Central American ◽

Significance Level ◽

Study Participants ◽

The Impact

Download Full-text

Impact of Prosthesis-Patient Mismatch on Survival after Mitral Valve Replacement: A Meta-analysis

The Thoracic and Cardiovascular Surgeon ◽

10.1055/s-0038-1675195 ◽

2018 ◽

Vol 67 (07) ◽

pp. 538-545 ◽

Cited By ~ 1

Author(s):

Ho-Young Hwang ◽

Suk-Ho Sohn ◽

Myoung-jin Jang

Keyword(s):

Mitral Valve ◽

Valve Replacement ◽

Odds Ratio ◽

Meta Analysis ◽

Early Mortality ◽

Term Survival ◽

Subgroup Analyses ◽

All Cause Mortality ◽

The Impact

Background Numerous studies have demonstrated a negative impact of prosthesis-patient mismatch (PPM) on long-term clinical outcomes after aortic valve replacement. However, the impact of PPM after mitral valve replacement (MVR) on clinical outcomes is still controversial. This study was conducted to evaluate the impact of PPM on early and long-term survival after MVR. Methods A literature search of five databases was performed. The primary and secondary outcomes were all-cause mortality and early mortality, respectively. Subgroup analyses were performed according to the risk of bias, patients' age, proportion of female patients, and proportion of patients with mechanical MVR. Results Eleven nonrandomized studies including 8,072 patients were included in this meta-analysis. The overall incidence of PPM was 58.0% (range: 10.4–85.9%). The odds ratio of early mortality in nine studies was not significantly different between the PPM and non-PPM patients (odds ratio: 1.35; 95% confidence interval [CI]: 0.98–1.86). A pooled analysis in 11 studies demonstrated that all-cause mortality after MVR was higher in the PPM than non-PPM patients (hazard ratio [HR]: 1.39; 95% CI: 1.09–1.77). This analysis revealed a moderate to high heterogeneity (I 2 = 69.4%). When pooled analyses were performed in two subgroups according to the proportion of patients with mechanical MVR, there were low heterogeneity in each group. No other subgroup analyses demonstrated a significant difference in the HR of all-cause mortality. Funnel plots and Egger's tests showed no visually and statistically significant publication bias. Conclusion The present meta-analysis indicates that PPM negatively affects long-term survival after MVR.

Download Full-text