Skip to main content

Questions tagged [meta-analysis]

Methods focused on contrasting and combining results from different studies, in the hope of increasing precision and external validity.

Filter by
Sorted by
Tagged with
4 votes
1 answer
276 views

Related low p-values that do not meet statistically significant thresholds

If I were to have multiple related measures, each with "low" p-values that do not quite meet the threshold of significance, could the fact that multiple measures produce a general trend with ...
Jaxton Winder's user avatar
0 votes
0 answers
20 views

Conducting a network meta-analysis in R - inputting dichotomous data

I am looking to conduct a meta-analysis in R using {netmeta}. However, my outcome of interest is a little bit complex and would appreciate advice in how best to enter it into the dataset. I am looking ...
emmz's user avatar
  • 1
1 vote
1 answer
35 views

How to find difference between two hazard ratios (ratio of hazard ratios)

I am reading a meta-analysis and in this paper the authors somehow compared HR from two groups with different treatments in each study. As I know, we can compare outcomes from two groups with ...
reress's user avatar
  • 11
3 votes
2 answers
111 views

Analyzing an experiment which consists of many "small" experiments

Consider the following setting: I am trying to choose between two bakers for my new bakery, Baker A and Baker B. The way that I am evaluating them is on a set of different cakes on the menu: Cake 1, ...
gaussiandynamics's user avatar
0 votes
1 answer
36 views

Robust Variance Estimation or Cluster Wild Bootstrapping on a multivariate meta-analysis

Cliffnotes Can I use RVE or cluster wild bootstrapping in meta-analytic models that are not meta-regression? Question I am conducting a three-level meta-analysis looking at the effects of a particular ...
Max Moser's user avatar
0 votes
0 answers
24 views

Conditional Logit Model - Utility Structural Estimation - Meta Analysis

I am performing a structural estimation of an utility function across several databases (from distinct articles) using McFadden (2001) framework (see reference). Each article's database includes N ...
RobM's user avatar
  • 1
1 vote
1 answer
79 views

Understanding correlated and hierarchical effects models in meta-analysis

I am doing a three-level meta-analysis, and I want to compare a three-level correlated and hierarchical effects model where effects are nested between-studies to a two-level model where variance at ...
Max Moser's user avatar
0 votes
1 answer
30 views

I was trying to calculate standard error from log odd ratio, Is this calculation correct?

I have two datasets. The first dataset have odd ratio, 95% confidence interval, p-value, and 2x2 table. while, the second dataset only have odd ratio, 95% confidence interval, and p-value without 2x2 ...
Epic Item's user avatar
1 vote
1 answer
34 views

Pooled Hazard Ratio or Crude events (M-H to RR or OR) for mortality outcomes in Meta-Analysis?

I'm performing a MA to estimate the effect of treatment A respect to Placebo (or propensity-matched control group) respect to all-cause death. All the studies included in my MA report both crude ...
Michele Maremmani's user avatar
0 votes
0 answers
19 views

Can I do a meta-analysis of categorical outcomes by assigning them to ordinal values in single-arm studies?

I have a collection of case reports and single arm cohort studies (without control groups) that describe the clinical response of patients treated with a particular drug. The response is reported ...
mortal1877's user avatar
0 votes
0 answers
41 views

Outlier and influential analysis for multilevel meta-analysis with robust variance estimation

I am conducting a three-level meta-analysis where I have different dependency structures in my data. I model the dependency by approximating the var-cov matrix, followed by estimating a three-level ...
user408318's user avatar
0 votes
0 answers
19 views

Is it possible to calculate effect size from the relative pre-post change?

I am trying to conduct a meta-analysis that focuses on the pre-post change of an intervention. While most studies reported the pre and post mean and SD scores of the outcome, some only reported the ...
Allie's user avatar
  • 1
5 votes
2 answers
188 views

Why are 1/SE or 1/variance commonly used as weights in regressions?

I'm trying to do a meta-analysis for the first time, comparing measurements of a simple experiment treatment against a control in a variety of species. I started by fitting a mixed-effects model to ...
mkt's user avatar
  • 18.9k
0 votes
0 answers
22 views

I2 (I-squared) estimates for heterogeneity of independent meta-analysis versus a single meta-regression

I am wondering how the I2 estimates of independent meta-analyses per level of a categorical moderator compare to those of a single meta-regression. The question is similar to the one about how the ...
befriendabacter's user avatar
1 vote
0 answers
23 views

Sample size calculation for prevalence studies: how do I find d when reviewing a published study?

I have found the following formula in Munn et al.: $$N = \frac{Z^2 P(1-P)}{d^2}$$ $N$ = sample size $Z$ = statistic for a level of confidence $P$ = expected prevalence $d$ = precision The problem I ...
Fabio Carraturo's user avatar
1 vote
0 answers
22 views

How can I convert effect estimates to be on the same scale? [closed]

I am doing a meta-analysis of studies that examine the effect of air pollution (continuous exposure) on a health outcome. Many of the studies report effect estimates for different units of exposure. ...
embracent's user avatar
0 votes
0 answers
21 views

How to meta-analyze when one condition has more replicates (within study) than another?

I am wondering how I can test, overall, does model X or model Y have greater prediction skill (pearson correlation coefficient)? It seems a meta-analytical approach is best suited, but is it possible ...
lost-gradStudent's user avatar
0 votes
0 answers
19 views

What is the standard deviation/uncertainty of a single observation in a meta-analysis?

I am running a simulation with parameters based on an existing random-effects meta-analysis. I would like to draw individual patients in the treatment group from a normal distribution with µ=the ...
bumble.bee's user avatar
0 votes
0 answers
20 views

Meta-analysis of odds ratio for continuous variables, how to interpret the decrease or increase per unit?

So, I am doing a meta-analysis of odds ratios to find out the risk factors for a certain medical outcome. One of those risk factors would be (simplified) the time spent in the operation room. My ...
echodoro's user avatar
4 votes
2 answers
58 views

Accounting for recall period in proportion meta-analysis

We are conducting a meta-analysis of "prevalence" of event-type exposures, e.g. "have you been physically assaulted in the past X months". We are trying to adjust for different ...
jessexknight's user avatar
2 votes
1 answer
100 views

Heterogeneity and ICC when using metafor's rma.mv and robust variance estimation

I am conducting a meta-analysis of standardised mean differences with many sources of dependency (multiple outcomes measuring the same construct, multiple control groups, outcomes from dyads (e.g., ...
AndreasV's user avatar
3 votes
1 answer
107 views

Forest plot to show results in a observational restrospective cohort study

I have to present my study as a poster and, in order to make my results more visual, I thought it was a good idea to report them as a forest plot. However, it is ok to use a forest plot to report the ...
user406570's user avatar
0 votes
1 answer
30 views

Can I extrapolate number of deaths (n_event) from 5-year survival Kaplan Meier?

Let's say there are 52 patients in the negative margin group and 24 in the positive margin group. I'm presented with this statement on the survival rates: "Overall survival was similar in the ...
Bilal Bahadır Akbulut's user avatar
0 votes
0 answers
18 views

Meta-analysis with effect sizes calculated from continuous and binary data [duplicate]

Can I compare effect sizes (ES) Cohen's g (from continuous outcome data) with effect size log odds (proportion outcome data) in a meta-analysis? Or can you only compare like for like, and why?
Max Moser's user avatar
0 votes
0 answers
83 views

Combining multiple outcomes/measurements in meta-analysis

I am conducting a meta-analysis with 12 studies. 4 studies use multiple ways to measure the same outcome or measure the same outcome at different time points. In any case, the outcomes/measurements in ...
FZZ's user avatar
  • 1
1 vote
0 answers
22 views

Balancing when combining multiple datasets (balancing in a metastudy?)

There have been some detailed discussion of balancing the datasets in this community: Are unbalanced datasets problematic, and (how) does oversampling (purport to) help? Why is accuracy not the best ...
Roger V.'s user avatar
  • 4,429
1 vote
0 answers
45 views

Meta-Analysis - Correctly treat multiple conditions in within study designs

we are currently conducting a systematic review and meta-analysis and are unsure how to deal with multiple effects from one and the same study. We are running a basic hierarcical three-level model ...
user408318's user avatar
1 vote
0 answers
21 views

Meta regression as mixed model: time interaction

In our meta-analysis, we have several potential moderators, both categorical (e.g. disease type, outcome type), and continuous e.g. publication year. Suppose we wish to investgate the effect of ...
user167591's user avatar
1 vote
1 answer
35 views

How to backtransform SMD to MD using metafor

I´m performing a prognostic effect meta-analysis of SMD in the metafor R package. Data includes means and SD for each arm (...
san festein's user avatar
6 votes
2 answers
78 views

meta-analysis: expected Q and observed Q

I understand the equation for observed $Q$: $$ \Large Q = \sum_{i=1}^k \frac{1}{V_i} (Y_i - M)^2 $$ But I don't understand why the expected $Q$ is just the degrees of freedom ($k - 1$), where $k$ is ...
confused meta-analyst's user avatar
0 votes
0 answers
20 views

How can I normalize Hedges' g from two extreme conditions around the control condition?

The variables I am discussing below are not the ones I am actually using but I think that they should give a better sense of what I am trying achieve. I am performing a meta-analysis of some chosen ...
aria's user avatar
  • 1
0 votes
0 answers
20 views

Meta-analysis of Kaplan-Meier survival estimates from one-arm studies

I am writing this topic because there is a statistical problem which does not let me sleep at night. For introduction, I am a clinical researcher interested in niche fields, and many of my projects ...
Mały Mi's user avatar
2 votes
1 answer
40 views

NA results after pooling estimates and coeff of mixed effects cox model from MICE imputation dataset

I need your help with my problem. So, after step of imputation missing data through MICE method, I got multiple imputed dataset. Then, I pooled the estimates and coefficients with mixed effect cox ...
Hoang-Giang Pham's user avatar
1 vote
0 answers
25 views

Converting confidence interval to SE for Ln(rate ratio)

I have the following a data for a medical meta-analysis. The data is given as an annualized flare rate for the drug and placebo, and then the rate ratio (0.67) is presented with a confidence interval. ...
Yomna's user avatar
  • 21
2 votes
1 answer
85 views

Rstudio metafor moderator analysis: one variable is missing in Model Results + eliminating intrcpt? [closed]

I am currently working on a meta analysis and right now I am trying to do a moderator/moderation analysis but I have come across one problem: When I try to run the function: ...
as07's user avatar
  • 21
1 vote
0 answers
43 views

Combining data before and after protocol modification

In one of my research, we decided to "update" the procedure mid-study. In short, we initially tested a repeated measures design (1 factor with 3 levels, 5 times per condition for a total of ...
Nee's user avatar
  • 11
3 votes
2 answers
84 views

Meta-analysis based on cohort studies only

I am planning a meta-analysis based on cohort studies only. There are dozens of papers reporting results of treatments A and B. But the treatments were never compared with each other. Besides, there ...
Viktor's user avatar
  • 1,035
0 votes
0 answers
78 views

Effect size + variance calculation single group pretest posttest - different sample size pretest-posttest

I include single group pretest-posttest study designs in my meta-analysis. I calculate the effect size as follows: difference between posttest mean and pretest mean divided by the standard deviation ...
FZZ's user avatar
  • 1
0 votes
0 answers
39 views

Multiple-comparison p-value adjustment in meta-analysis

Do we need to adjust $p$-values for subgroup analyses in a meta-analysis? I'm wondering if the typical subgroup analysis methods, like the ones in packages such as ...
tomas.rosa's user avatar
2 votes
0 answers
53 views

Multivariate meta-analysis with sparse multivariate array

I am wishing to conduct a meta-analysis that has many end-point measures that are all supposed to reflect a single concept - let's call it poor mental health. That would include things like ...
J Taylor's user avatar
  • 401
5 votes
2 answers
179 views

How can I calculate the weight of a study for meta-analysis?

I'm tasked with producing a forest plot for a meta-analysis but I've only been given the risk ratio, sample size and confidence intervals of the necessary studies. I think that I have to find the SE ...
user356816's user avatar
2 votes
0 answers
74 views

Comparing Multi-level Meta-regression to Network Meta-Analysis for Dose-response Project

I am planning a meta-analysis to examine the average change in a continuous variable over time within my area of research. To be able to meta-analyze this continuous relationship, I've typically ...
zr2015's user avatar
  • 41
6 votes
3 answers
971 views

Do 95% Confidence Invervals use +- 1.96 standard error or standard deviation?

I'm trying to calculate the weight of studies for inclusion in a meta analysis as part of homework. I've been given a number of studies and their final results represented as a risk ratio with 95% ...
user356816's user avatar
0 votes
0 answers
21 views

How can I combine fully measured with partially measured outcomes in an IPD (Network) Meta-Analysis?

I was asked to help in an Individual-Patient Data (IPD) Network Meta-Analysis (NMA). Outcomes are supposed to measure combined scores of condition 1 and condition 2 (thus: $Score=Score_1 + Score_2$), ...
Federico Tedeschi's user avatar
3 votes
1 answer
34 views

Calculate SE of regression coefficients using p-value (for meta-analysis)

I would like to do a meta-analysis of the regression coefficients from a group of 9 studies, but they authors had not reported the SEs. For all of the studies, the tables included in the reports only ...
plainsong's user avatar
0 votes
0 answers
44 views

Meta-analysis: estimating partial correlation coefficients using estimates from multi-level analyses

I would be very grateful if someone could offer some ideas or point to relevant literature on the calculation of partial correlation coefficients (PCCs) using estimates from multi-level analyses. ...
Mugen's user avatar
  • 1
0 votes
0 answers
17 views

Converting from Chi Square to Hedges in the context of mixed effects model comparisons

I'm attempting to extract a Hedges' G effect size from a publication that reports the chi square statistic (and degrees of freedom) derived from comparing two mixed effects models: the first model ...
Victor Pokorny's user avatar
0 votes
0 answers
26 views

SE and CI not aligning in R output from RVE meta-regression (robumeta)

I implemented a random-variance-estimation meta-analysis in R using the robumeta command. I am puzzled by the R output in terms of interpreting the significance level. I am copying an example of the ...
Janina Steinert's user avatar
0 votes
2 answers
52 views

Testing publication bias and heterogeneity in meta-analysis when using multiple studies

I am conducting a meta-analysis and I'm trying to ascertain how to test for publication bias and heterogeneity. However, I have used multiple datasets to calculate effect sizes, and therefore I'm ...
User124444's user avatar
2 votes
1 answer
70 views

month level fixed effects model

I am trying to understand the effect of migration of a male member on the employment outcomes of left-behind women. I have monthly observations for 5 years. I am using a fixed effects model with month ...
sapna's user avatar
  • 21

1
2 3 4 5
30