Network Meta-Analysis for Beginners: Concepts, Steps, and Practical Insights

Name: Network Meta-analysis for Beginners in 25mins
Uploaded: 2026-01-15T07:55:23.572090+00:00
Channel: Innocent David
Description: Summary and key takeaways on Network Meta-Analysis for Beginners: Concepts, Steps, and Practical Insights, covering Introduction The video provides a

Innocent David

Jan 15, 2026

•

4 min read

YouTube video ID: Ouuq6CIooVs

Source: YouTube video by Innocent David — Watch original video

PDF

Introduction

The video provides a crash‑course on network meta‑analysis (NMA), aimed at beginners who already know the basics of systematic reviews and traditional meta‑analysis. It focuses on theory and foundations rather than hands‑on coding, with future videos promised to cover practical implementation in R.

What Is Network Meta‑Analysis?

An advanced form of meta‑analysis that compares multiple interventions simultaneously.
Uses a network of nodes (each node = a treatment) and edges (lines) that represent direct comparisons between treatments.
Allows direct (studies that compare two treatments head‑to‑head) and indirect comparisons (e.g., A vs C inferred through a common comparator B).

How NMA Differs from Traditional Meta‑Analysis

Traditional Meta‑Analysis	Network Meta‑Analysis
Compares two interventions (A vs B).	Compares many interventions (A, B, C, …) at once.
Relies only on direct evidence.	Integrates direct and indirect evidence to create a complete picture.
Provides a single pooled effect.	Generates a treatment ranking and relative effects for every pair of interventions.

Why Use Network Meta‑Analysis?

Overcomes the limitation of pair‑wise comparisons by handling complex treatment networks.
Helps identify the optimal (most effective or safest) treatment among several options, which is valuable for clinical guidelines and policy decisions.
Produces more comprehensive evidence by borrowing strength from indirect comparisons.

Key Concepts

Nodes: Represent individual treatments or interventions.
Edges (Lines): Show direct comparisons; thicker lines indicate stronger evidence (narrower confidence intervals or more studies).
Node Size: Larger circles reflect a greater number of studies involving that treatment.
Direct Evidence: Data from trials that directly compare two treatments.
Indirect Evidence: Inferred comparisons via a common comparator (e.g., A vs C through B).

Core Assumptions

Transitivity – The studies being combined should be clinically and methodologically similar (patient characteristics, interventions, outcomes). This permits valid indirect comparisons.
Consistency – Direct and indirect evidence for the same comparison should agree. Inconsistency may signal bias or heterogeneity.

Visualising the Network

Network Plot: A graphical summary showing nodes, edges, node sizes, and edge thicknesses.
Interpretation Tips:
Larger nodes = more studies for that treatment.
Thicker edges = more precise (stronger) direct evidence.
Gaps in connections highlight where indirect evidence is crucial.

Main Outputs of an NMA

Relative Treatment Effects (e.g., odds ratios, mean differences) for every pair of interventions.
Treatment Ranking – Ordered list from most to least effective based on a chosen outcome.
SUCRA (Surface Under the Cumulative Ranking Curve) – Quantifies the probability that a treatment is among the best.
League Tables – Pair‑wise comparison matrix showing effect estimates and confidence intervals.
Network Plot – Visual representation of the evidence structure.

Step‑by‑Step Workflow

Define the Research Question using the PICO framework (Population, Intervention, Comparator, Outcome).
Systematic Literature Search across databases such as PubMed, Embase, Cochrane CENTRAL, Scopus, ClinicalTrials.gov, etc.
Data Extraction – Collect treatment arms, outcomes, sample sizes, and study characteristics for each eligible trial.
Build the Treatment Network – Map direct comparisons to visualise the network structure.
Run the NMA Model – Fit a statistical model (frequentist or Bayesian) to estimate relative effects and rankings.
Assess Assumptions – Check transitivity, consistency, and heterogeneity.
Present Results – Use network plots, league tables, SUCRA values, and ranking plots.

Software Tools

R (with packages netmeta, gemtc, meta, rjags for Bayesian models).
RStudio – User‑friendly IDE for running R scripts.
STATA – Commands like network and mvmeta.
RevMan – Primarily for pair‑wise meta‑analysis; limited NMA capabilities.
WinBUGS/JAGS – For Bayesian NMA implementations.
Cytoscape/Gephi – Optional for custom network visualisation.

Take‑Home Messages

NMA extends traditional meta‑analysis by allowing simultaneous comparison of many treatments, providing a holistic view of efficacy and safety.
It relies on direct and indirect evidence, demanding careful assessment of transitivity and consistency.
Visual tools (network plots, node/edge sizing) help interpret the evidence structure.
The typical workflow mirrors that of a systematic review, with the added step of network construction and model fitting.
Open‑source software, especially R, offers powerful, flexible packages for conducting NMAs.

Looking Ahead

Future videos will demonstrate the practical side: importing data into R, constructing the network, running the model, and interpreting outputs such as SUCRA and league tables.

Network meta‑analysis transforms multiple‑treatment evidence into a single, coherent framework, enabling researchers and clinicians to pinpoint the most effective option while accounting for both direct and indirect data—making it an essential tool for evidence‑based decision‑making.

Frequently Asked Questions

Who is Innocent David on YouTube?

Innocent David is a YouTube channel that publishes videos on a range of topics. Browse more summaries from this channel below.

Does this page include the full transcript of the video?

Yes, the full transcript for this video is available on this page. Click 'Show transcript' in the sidebar to read it.

What Is Network Meta‑Analysis?

- An **advanced form of meta‑analysis** that compares **multiple interventions simultaneously**. - Uses a **network of nodes** (each node = a treatment) and **edges** (lines) that represent direct comparisons between treatments. - Allows **direct** (studies that compare two treatments head‑to‑head) and **indirect** comparisons (e.g., A vs C inferred through a common comparator B).

How NMA Differs from Traditional Meta‑Analysis

| Traditional Meta‑Analysis | Network Meta‑Analysis | |---------------------------|----------------------| | Compares **two** interventions (A vs B). | Compares **many** interventions (A, B, C, …) at once. | | Relies only on **direct** evidence. | Integrates **direct and indirect** evidence to create a complete picture. | | Provides a single pooled effect. | Generates a **treatment ranking** and relative effects for every pair of interventions. |

Why Use Network Meta‑Analysis?

- Overcomes the limitation of pair‑wise comparisons by handling **complex treatment networks**. - Helps identify the **optimal (most effective or safest) treatment** among several options, which is valuable for clinical guidelines and policy decisions. - Produces **more comprehensive evidence** by borrowing strength from indirect comparisons.

Summarize another video

Full Transcript YouTube

Hey guys, what's going on there? In this
video, I'm going to be talking about
network meta analysis for beginners.
So, if you look at my past video, I've
made a lot of videos on meta analysis,
how to conduct systematic reviews and
also videos around several domains. Now,
in this particular video, I'm going to
be talking about what meta analysis is.
And now one of the things I'm not going
to be doing in this video is I'm not
going to really practice the research
question because I actually want you to
understand the foundation and have a
very theoretical understanding of what
network meta analysis really is. Now in
my subsequent videos I'm going to
actually employ practical tools using R
studio on R you know to execute
practicals on how to perform a network
meta analysis. So actually enjoy you to
like this video, share this video and
actually engage and tell me areas you
would want me to make further future
videos around. Now let's get started.
Now we are going to be talking about
network meta analysis once more for
beginners. So we're going to actually be
looking at some of the concepts and some
of the practical steps and uh one of the
things we need to actually understand is
that we might want to know what is a
meta network meta analysis.
Now, usually beyond the normal
traditional meta analysis that we
usually do where we pick one
intervention A versus intervention B and
we try to actually draw comparisons
between these two interventions to get
uh an estimate that would could say that
Suzu and so favors this intervention
over the other intervention. Okay. So I
just kept that simple.
Now when we talk about network meta
analysis, now what changed from meta
analysis? One of the things that changed
is that we use the word network in it.
So it's more like actually an advanced
form of meta analysis that allows for
more comprehensive comparison. Now when
we say networks now think about it as a
node. You have a node here. So we've got
node A, got node B, we've got node C,
got node D and E. Now between these
nodes, now there are several connections
between node A and B. Is the connection
between node A and C is a connection
between node A and E is a connection
between node A and F or depending on how
many nodes there. So what network
analysis tries to understand is that we
want to actually compare multiple
interventions or want to compare
multiple interventions right either
directly or indirectly. So here when we
talk about multiple treatments one
analysis essentially what this means is
that it it simultaneously compares you
know multiple treatment even those that
are not directly tested against each
other. So it can't be that we're trying
to compare paracetamol versus metronid.
So basically what I actually want you to
understand here now is that when we are
comparing two interventions right maybe
we're comparing one drug A versus drug B
and drug C. Now it could be that we're
trying to compare drug A and B or maybe
job A and C but invariably we still see
ourselves showing estimate of job A
versus B which is what more like an
indirect what comparison. So whenever we
are performing a meta analysis more like
we have one research question then we
now want to compare several
interventions. So it leverages all
evidences right. So in network meta
analysis, it integrates both direct uh
you know and indirect evidence just like
I explained to form a complete picture.
So just think about it as you want to
compare a certain group of drugs about a
particular condition and want to
actually know which of them that is most
efficacious amongst all of them. Now the
traditional meta analysis is usually
done between what two interventions
where we try to compare between class A
versus class B. But in a network meta
analysis we can compare multiple
interventions at the same time either
directly or indirectly. So another thing
we need to actually understand about
network meta analysis is that we are
trying to identify the optimal
treatment. Okay. So the the goal is to
actually understand which of them which
is most efficacious or effective or
probably most treatment amongst several
options. So it's actually a very massive
area for you to get started with.
Now why do we need network meta
analysis? Okay, that's the question. Now
looking at the first explanation that I
did make. Now you would understand that
some of the limitations of the
traditional meta analysis is that we are
only comparing two treatments at a time.
Now but in terms of a network meta
analysis, we actually trying to address
some kind of complexities between
several treatments at the same time. So
it means that any research question that
that is being drawn from the position or
the statistical standpoint of a network
and meta analysis is trying to
understand the best approach across a
wide array of interventions or a wide
array of treatments. So it's actually
very important for you to understand the
foundations of network meta analysis
before we dive into the practicals
because the network meta analysis makes
uh it makes the process invaluable for
developing what clinical guidelines
because at the end of the day what we
might say is that we are trying to
understand which drug is most effective
for managing type 2 diabetes and we have
metforming we have uh maybe another drug
We have another one for another drug. We
have another one that's class four
drugs. Now essentially what this
particular procedure is doing is that we
want to now know which of which amongst
all of these four drugs that is
currently on the treatment pathway is
most efficacious or is more safe
relatively safe you know for policy
makers to place probably at the first
line of you know uh treatment pathway or
second line depending on how the the
number of nodes that we are trying to
compare.
Now I'm going to actually explain
something very simple in this crash
course because the idea for this course
is to make it very simple and easy for
each and every one of you to understand
what we are talking about. Now just like
I actually explained about having
multiple nodes
when we talk about the meta network meta
analysis now it is not really farfetched
from the traditional meta analysis where
we have two interventions but here is
the point that I actually want you to
get right away that when we are
comparing or when we are performing a a
direct meta analysis or a traditional
meta analysis now here are some of the
options we might have might actually
have be treatment A versus treatment B
or have treatment A versus placebo or
have treatment A versus control or have
treatment A versus another form of
intervention. And now what type of
intervention is it? This is more like a
direct comparison because we are trying
to make inference from drug A versus
drug B with respect to some relative
outcomes of interest. So this could be
some kind of form of primary outcomes
like maybe alltime mortality. It could
be maybe uh change in baseline uh blood
pressure or change in baseline glucose
level whatever intervention or outcome
that you are targeting. But the the good
stuff of this explanation is that when
you're conducting a normal traditional
meta analysis, it usually involves two
interventions, right? Got class A and
class B and then you do a direct
intervention. Now, but let's look at
what a network meta analysis looks like.
Here is that you can actually be doing a
direct evidence, right? where you're
comparing node A versus node B, right?
Where you're trying to actually
understand how directly connected this
particular class of drug is to this
particular class of drug, right? Or it
could be any form of intervention aside
from uh therapeutics or whatever.
Now, but the idea here is that when you
talk about direct evidence, you're
saying, okay, we're trying to make a
comparison of treatment A versus B
directly, maybe two randomized trials.
But if we actually looking at indirect
evidence now since we're comparing node
A with node B. So indirectly we can
actually be having pulled evidence of
node A to node C. Now that's why we say
it's a comparison of treatment of A
versus C which is actually inferred
through a common comparator of B which
can be A versus B, B versus C which also
implies A versus C. Okay. So these are
actually some kind of indirect evidence
that pulls on when you're actually
performing a network meta analysis and
and I'm going to talk about some of the
strengths of a network meta analysis
because uh when we actually have the
opportunity of comparing multiple
interventions at the same time uh
essentially what it does is that uh it
generates a more comprehensive treatment
evidence effect estimate for us. So uh
on the course of interpreting the
results from a network meta analysis,
what it does is that you you actually
get much more uh you know solid evidence
which is gotten from d indirect uh you
know studies that may or direct indirect
probably interventions that wen't of
interest to you but you already have the
result pulled from there. So you can
actually now make inference from it into
your what your result.
Now the next thing we're going to
actually look at now is I'm just going
to paint a visual picture of what a
network looks like because when I talk
about network meta analysis couple of
persons might be lost with what I was
talking about but basically this is what
a network looks like okay now what's a
network now a network is comprises of a
node right a node so when we talk about
a node now is that we've got several
nodes we've got node A here we've got
node B we've got node C and uh till node
I whatever nodes right so the treatments
uh some of them were actually the nodes
this is node A this is node B this is
actually another node C so when talk
about nodes we're talking about several
treatments and uh each circle each of
these nodes like I said they represent a
treatment which is going to be compared
now [snorts] we're going look at what we
call the lines now and what do we mean
by lines and when you actually look at
some of these nodes here you see that
there are actually some lines that are
connecting each of them you see line
node A is being connected to node B and
node B is connected to node C. Node C is
connected to node E and that of that.
This is what purely a network is all
about. Now think about this network as
MTN or think about it as any kind of
traditional networking system or
networking package that's in your
country and when you look at some of
these nodes or when you look at some of
these networks or you see that the
networking when you've if if you're
somebody that that you you've done
computer networking you see that the
internet is connected with various
nodes. So look at it that your phone is
connected to the server out there
online. The server is connected to
another phone. That phone is connected
to another server that is also connected
to another Wi-Fi is connected to several
nodes. Now that is what we talk about.
Now what connects them is what we call
the line. And this is a very very
crucial part in the interpretation of
our network meta analysis. And that's
why it's actually very imperative for
you to understand
what this picture is and actually create
a mental understanding of what we're
talking about. So the lines are more
more like here that you can actually see
at the edges they connect the treatments
you know that have been directly
compared in a study. So what it means
that when we look at study A and study B
we see that what connects them is this
line connecting two of the studies which
or these interventions could actually be
serving as a node. So they have line
lines connecting them. We're actually
going to actually see what we call the
visual cues. And now what do we mean by
visual cues? So actually say that
sometimes some larger notes often
signify more studies you know for that
treatment. Why thicker lines you know
indicates stronger evidence for that
comparison. Now I'm actually going to
actually explain this as think about it
as whenever we're trying to interpret
the diamond when we are doing our
traditional two two-way meta analysis
where we're comparing two interventions
or two treatment effect. Now when when
the diamond when the when the diamond is
big what do we say we can actually say
there and maybe probably there is a more
wider confidence interval or there's
more study weight in which one study has
more weight compared to the other study
and essentially what this means is that
when we inferring this into network meta
analysis that has several interventions
connecting uh to them to several nodes
it's trying to explain in simple details
that larger nodes now will often signify
more studies is for that particular
treatment. Do you understand? Now why
some of the thicker lines why there is
now thicker lines which we can actually
interpret as usually confidence
intervals whenever we are doing that uh
direct study comparisons for two
studies. So those thicker lines are now
going to indicate stronger evidence for
that comparison. So this is just like a
mental picture that I want you to have
based on this particular network image
that I have displayed right here.
Now for the network plots. Now visually
they represent the available evidence
you know the available evidence
structure you know making it easier to
understand which treatment have been
compared and how robust you know this
evidence is.
Now we're actually going to look at
something that we call some of the
assumptions in network meta analysis.
And what are some of the assumptions
that we have in network meta analysis?
Some of them is transitivity. You know
most times there is this assumptions
that studies are clinically and
methodologically similar you know enough
to allow for indirect comparisons and uh
this is something that means that you
know the characteristics of the patients
and interventions across different
trials should be comparable. So it
doesn't mean that uh you should not make
inference you know when uh there are
some kind of hetrogenity in the trials
that you're trying to compare in that
pool of network intervention or network
meta analysis. And uh another one there
is what we call consistency right so
many of times you know there's this
assumption that direct and indirect
evidence for a particular comparison
provide similar effect estimate
and these inconsistencies can indicate
underlining issues in a network and one
of the things I want you to understand
is that most times just like we have for
the normal traditional meta analysis
where we do account for homogenity
across some of the trials you know uh it
is essential for you to understand that
trials included in the talk you know
some of them should have similar patient
population and this is an assumption
that is usually being made outcome
definitions intervention duration and
there are several parameters that we
usually extract during our data
extraction more like I would say study
methodologies to ensure valid
comparisons
[snorts] so it is important for you to
know that if you violate some of these
assumptions there are tendencies for it
to lead to bias results you know making
careful assessments crucial for validity
of the network meta analysis And I think
one of the things you need to understand
from this very short crash course is
that when I'm I'm actually
[clears throat] going to pull out a
practical on this uh essentially this is
actually going to make a lot of sense
because it will give you uh some of the
highlights and also how to understand
this uh much more properly and clearer.
[snorts] Now what are actually some of
the main outputs for a meta analysis?
Now I would say for a normal meta
analysis we look at confidence interval
we look at some of the uh we look at
some of the hetrogenity score which is
usually measure t squ that's in between
study consistency as trying to measure
internal consistency we've got uh
several parameters like the I squared
when we're trying to understand
hetrogenity among studies and uh our
pool effect and all of that we've got
forest plot and all of that funnel plot
as usual now but in the case of the
network meta analysis. Right now we have
what we call a network plot. So that is
what we are also going to generate on
the cost of our practicals. We're going
to generate a network plot and
essentially what the network plot is is
that it's actually a visual
representative of the treatment
comparison and available evidence.
Now another parameter that we are going
to be interested in is what we call the
treatment ranking. And you know just
think about this more like as a
hierarchal you know ordering of
treatments from best to worst based on a
specific outcome. you know often
visualize using ranking you know
probability. So that's what we use the
ranking probabilities to show whenever
we display the results of a network meta
analysis. [snorts] Now another thing
we're going to that's also going to be
important to us on the course of this
course is a relative treatment effect.
Let's think about it as the overall
effect when you're when you're doing a
normal traditional meta analysis. So in
this case uh we're going to actually be
looking at some of the estimates of how
each of these treatments you know
compares to every other treatment in the
network because we have several nodes
that are being connected with several
lines. Remember how I actually explained
it in my previous slide. Now but what we
want to know now is I want to actually
understand what's actually the estimate
of how some of these treatments compares
to every other treatment that's on that
node in the network and how is more many
times you actually present this as you
know all ratios you know mean
difference. So some of those statistical
estimates that I have actually explained
to you in my previous videos for a meta
analysis also applies in your network
meta analysis for the interpretation of
a resource. Now another thing that we
need to actually understand is what we
call the leak tables. Okay. So the leak
tables and what we call the sucra uh so
more more like what the leak tables does
in the network meta analysis is that it
provides a comprehensive pair wise
comparison. Right. So why uh the surface
under the cumulative ranking score which
you call the sucra uh ranking curve
rather what it does is that this this
act this ones quantify the probability
scores uh of a treatment that is being
among some of the best. So we're
actually going to explain all of this
you know show you how the nick tables
works in network meta analysis and how
you can actually get started at least
with the basic meta analysis of a pool
prevalence across several studies.
Now we're actually going to be looking
at some of the steps for you to perform
a meta analys network meta analysis. So
now this is actually not different from
what we do in systematic review or a
meta analysis where you start for a
network meta analysis you define your
research question. Now it's actually
very very important for you to
understand that whenever you're
performing a network meta analysis it's
usually best to whenever you're using an
effect of intervention estimate and a
research model or research question that
is usually suitable for this is what we
call the pico model and what's the ple
model what the ple model does is that
you have your population you have
intervention you have a comparator you
have your outcome so we we try to
clearly specify what is the population
intervention comparator and outcome. So
we actually pick out all these
parameters because they're actually very
important when we are queuing in our
data. And then essentially the next
thing you're going to do now is what
you're going to do is systematically
search. So that means you're going to
actually search on several databases
like concrete central which is actually
one of the best informed database for
you to identify trials and you can
actually search at trial.gov. You can
also search on you know mbase via ovid
can search on you know medline via ovid
you can also search on scopus you can
search on pubmed and there are a lot of
repositories out there that is going to
be very important for your search. Now a
very crucial part of our network meta
analysis step is what we call the data
extraction. So what do you do by data
extraction? It means that you're
definitely going to pull out studies
from all of these findings that you've
pulled out. So it means that if you've
got like nine trials, you have to
extract stud information parameters from
each of them because this quantitative
estimates that you're going to extract
from them is going to be very relevant
for you to uh queue it in for us to
perform a network meta analysis. And
that means you have to extract all the
key information from included studies
including treatment arms, you know, the
out outcomes, you know, probably primary
outcome, secondary outcomes that are
going to be very imperative for your
exploration.
And then also uh you can also build the
treatment network which is something I'm
going to show you in the next video in
my practicals and probably how to map
out the direct comparisons to visualize
the network structure and then we're
going to run the network meta analysis
model. That's definitely going to be the
last thing that we're going to do. Okay.
So remember the first thing to do is to
define your research question. You do
your systematic decision research for
your effect of intervention uh meta
analysis research question. Then you
extract the relevant data you know uh in
several treatment arms that you're going
to actually queue in. Then we build a
treatment network which is what I'm
going to show you in practicals by
mapping out some of the direct
comparisons to visualize the network
structure. Then we run the network meta
analysis model.
Now the the most important thing there
is that as someone that really really
really likes review manager I've used
review manager severally for several
meta analysis but uh in terms of a
network meta analysis uh there's a lot
of limitations there but there actually
a lot of packages software that are
quite very handy for you to start your
network meta analysis and one of them is
what we call the R software now when you
get install your R you also need to
install the R studio which is going to
make give you a much more better
interface for you to run your network
meta analysis and some of packages that
are going to be very imperative for you
to do that. You we've got the netmeta,
we've got the GMTC packages and lots of
them. Okay, so these are actually
powerful opensource statistical
programming language, you know,
dedicated packages for network meta
analysis. Also, we've got start. Start
is also very important, a very good uh
tool uh that is usually used for medical
research, also social sciences and all
of that. So, it's actually very
important for network meta analysis.
Other ones that we've got is Revman but
there are a lot of limitations because
uh Revman interface supports you know a
pair wise meta analysis and whenever I
actually see a pair wise meta analysis
talking about you know two meta analysis
that trying to compare one particular
intervention with another right compared
to a network meta analysis where you've
got several you know interventions that
you are comparing either directly or
indirectly.
Now also another tool there is we've got
bioam frameworks you know uh like the
wind box you know quite very important
and uh yeah so basically uh these tools
lots of them uh they really make it very
easy you can actually check to like
Julius airspaces but uh for much more
compatibility I would always really
recommend our software for that or start
now so what actually some of the key
take home messages that I want to leave
you with right now is that I just want
you to understand that network meta
analysis is all about uh you know is
more like simultaneous comparisons of
multiple treatments just like I
explained we are trying to compare more
than one more than two interventions at
the same time you know directly or
indirectly to provide a kind of a
holistic view of treatment efficacy and
uh it's something that optimizes
evidence uh just think about it that
when we study a certain condition and
we're just limited to comparing two
drugs at the same time uh it it it just
limits uh the amount of evidence we can
pull out from there. But compared to
when we have a network meta analysis
network array of drugs and uh we we are
already comparing several of them to
each other it gives us that particular
platform to understand which of them is
best or which of them has the best
performance on the treatment pathway
compared to single or double pair wise
you know meta analysis. Now also it
guides decision making essentially from
that it's going to lead to decision
making clinicians we know which study is
more efficacious and uh yeah that guides
decision making by making making
everything much more easier and uh for
clinicians to actually get gain robust
evidence. Now another one now is uh
validity uh you know assumption
dependent uh you know it gives us
validity which is one that hinges on
critical assumptions like transitivity
consistency which require careful
evaluation. Now the next thing like I
always do advise is practice and uh
essentially when you actually engage in
p practice it's uh it gives you much
more realtime knowledge practical
applications and real world examples of
what to do and uh a clear pathway of how
to actually get started with it. So I I
think I actually want to use this
opportunity to draw close to this
particular crash course and uh uh
introduction to network meta analysis
and I'm going to actually providing much
more videos on how to get started on
network meta analysis practically where
I'm going to show you run you through R
then from there we'll actually get
straight to action. So I actually want
to say very big thank you for your time
and uh have a great day. Bye-bye.