MetaLens v13.3 — Meta-Analysis Studio by Bayu Satria Wiratama

📋 Project Information

Project / Review Title

Research Question (PICO/PICOS)

Review Type

Date

📏 Analysis Configuration

Statistical Model

Both models are always computed and reported side by side; this choice sets which one leads the results.

Between-study variance (τ²) estimator

Hartung–Knapp adjustment

Confidence Interval Level

Reporting scale for pooling

Studies reported on other metrics are converted to this scale at analysis time. Your data table is never modified.

Prevalence Transform

🎛️ Custom Moderator Variables — for subgroup and meta-regression

💡

Choose your study type, then import via Excel/CSV or enter studies manually. Each study type has its own dataset.

📂 Select Study Type & Import

Study Type to Import

📂 Click or drag & drop Excel / CSV

Supports .xlsx, .xls, .csv

The workbook contains one sheet per analysis type, a data dictionary and a worked example row.

✏️ Add Single Study Manually

🎯 Effectiveness (0)

🩺 Diagnostic (0)

📉 Prevalence (0)

🔗 Association (0)

📭

No studies yet for this type.

🔄

Convert raw statistics into effect sizes. Use "Add to Dataset" to send the result directly into the correct analysis module.

Means & SDs

2×2 Table

Correlation r

t / F stat

d ↔ r ↔ OR

Any effect size

📊 Two-Group Means & SDs → SMD / MD

Intervention Group

Mean (M₁)

SD (SD₁)

N₁

Control Group

Mean (M₂)

SD (SD₂)

N₂

SMD Method

📊 2×2 Table → OR, RR, RD, NNT

Events – Intervention (a)

No Events – Intervention (b)

Events – Control (c)

No Events – Control (d)

📊 Correlation r → Fisher's z

Pearson r

Sample size N

📊 t-statistic / F-statistic → d

Statistic type

Value

N₁

N₂

📊 Cross-measure: d ↔ r ↔ OR ↔ logOR

Input measure

Value

N (for variance, optional)

n₁:n₂ ratio (default 1)

⚖️ Convert between any two effect-size metrics

Uses the same conversion engine as the harmoniser, including delta-method variance propagation.

Reported metric

Value

Variance

Standard error

95% CI lower

95% CI upper

Convert to

n (intervention)

n (control)

Pooled SD

Control event risk

💡

Generates a comprehensive report across all study types and analyses run in this session.

📥 Downloads

👨‍🔬

Bayu Satria Wiratama, Ph.D

Assistant Professor

Department of Biostatistics, Epidemiology, and Population Health

Faculty of Medicine, Public Health, and Nursing (FK-KMK)
Universitas Gadjah Mada, Yogyakarta, Indonesia

✉bayu.satria@ugm.ac.id | 🏛fkkmk.ugm.ac.id

📖 About MetaLens

MetaLens is a free, browser-based meta-analysis studio developed for researchers and students in health sciences. It supports the full meta-analysis workflow — from data import and effect size conversion to pooling, heterogeneity testing, publication bias assessment, subgroup analysis, and meta-regression.

The tool handles four study types in a single workspace: Effectiveness (RCTs and interventions), Diagnostic accuracy (sensitivity/specificity/SROC), Prevalence (Freeman-Tukey/logit transforms), and Association (observational correlations and ratios).

All computations run entirely in your browser — no data is ever sent to any server. Your study data stays on your device.

📋 Citation

Wiratama BS. MetaLens: a browser-based meta-analysis studio. Version 11.0 [software]. Yogyakarta: Department of Biostatistics, Epidemiology, and Population Health, FK-KMK, Universitas Gadjah Mada; 2026. Available from: http://ugm.id/metalens

Please acknowledge this tool in your publications when using it to conduct or assist your meta-analysis.

⚖️ License & Terms

MetaLens is released for free academic and research use. Redistribution with attribution is permitted. The software is provided "as is" without warranty. Always verify statistical results independently for publication.

For questions, bug reports, or feature requests, contact bayu.satria@ugm.ac.id.

📊 Statistical Methods

Analysis	Method
Pooling (RE)	DerSimonian-Laird
Pooling (FE)	Inverse-variance
Heterogeneity	Q, I², τ², H²
Publication bias	Egger, Begg & Mazumdar, FSN
Subgroup	Between-group Q (Qb)
Meta-regression	WLS, tau² null model
Diagnostic	Moses-Littenberg SROC, logit RE
Prevalence	Freeman-Tukey arcsine
Prediction interval	Higgins–Thompson–Spiegelhalter (t, k−2 df)
Small-study adjustment	Duval & Tweedie trim-and-fill (L₀ / R₀)
Sensitivity	Leave-one-out, cumulative, influence diagnostics
Sequential monitoring	TSA — O'Brien–Fleming α-spending, D²-adjusted RIS
Certainty of evidence	GRADE (8 domains, Summary of Findings)

🔒 Privacy & Data

All study data you enter is stored only in your browser's localStorage — it never leaves your device. No analytics or tracking are added by this app.

Project Setup

Bayu Satria Wiratama, Ph.D