<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@4eb77ebd8ef744eba5992a8cb85fec93" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Introduction</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@e278e9e5e19e41dc9cfd0bcb30791f3e">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@e278e9e5e19e41dc9cfd0bcb30791f3e" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p><strong style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-style: inherit; font-variant: inherit; font-stretch: inherit; line-height: 1.4em; font-family: inherit; vertical-align: baseline;">Introduction</strong></p>
<p>Welcome to the Exploratory Data Analysis (EDA) workshop. In this module, we build on the concepts covered in Section 2.07 and walk you through the Rmarkdown tutorial introduced in that section. We will explore techniques on how to display and prepare your data for further analysis. </p>
<p></p>
<p style="text-rendering: optimizelegibility; margin-right: 0px; margin-left: 0px; padding: 0px; border: 0px; outline: 0px; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-stretch: inherit; font-family: 'Open Sans', 'Helvetica Neue', Helvetica, Arial, sans-serif; font-size: 18px; vertical-align: baseline; color: #313131;"></p>
</div>
</div>
<div class="vert vert-1" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@ac1cc14c4423446e8e084ade7a8c0187">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@ac1cc14c4423446e8e084ade7a8c0187" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p></p>
<p><strong>Learning Objectives</strong></p>
<p>1. Learn how to summarise your data.</p>
<p>2. Understand how to create publication-ready layouts and learn the R package <em>tableone</em></p>
<p>3. Get to know different tools to create variables and plot their relationships</p>
</div>
</div>
<div class="vert vert-2" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@cfd9860b307841429aa7886e96584f9f">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@cfd9860b307841429aa7886e96584f9f" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p><strong>Credit</strong></p>
<p>Workshop Author:<span style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-style: inherit; font-variant: inherit; font-weight: inherit; font-stretch: inherit; line-height: 1.4em; vertical-align: baseline;"> Jesse Raffa</span></p>
<p></p>
<p><span style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-style: inherit; font-variant: inherit; font-weight: inherit; font-stretch: inherit; line-height: 1.4em; vertical-align: baseline;">Adaptation: Grigorij Schleifer and Louis Agha-Mir-Salim</span></p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@0298a725c1ec4a64a99f1b195d0a03b6" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Principles of Exploratory Data Analysis (EDA)</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@d17289b1776c4895aa142a5cfa1cb0c8">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@d17289b1776c4895aa142a5cfa1cb0c8" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>EDA’s goal is to better understand the data and the process by which it was generated.</p>
<p>Within statistics, it is largely considered separate from inferential/confirmatory statistics (e.g., hypothesis testing, point and interval estimates, etc), where EDA has a very diverse and important set of goals:</p>
<ul>
<li>Provide an opportunity to do additional data cleaning.</li>
<li>Understand how the data is generated, and what the relationships between variables may be.</li>
<li>Suggest questions and hypotheses that can be subsequently answered and tested.</li>
<li>Identify what statistical methods may be most appropriate for the data to follow up with these questions and hypotheses.</li>
</ul>
<p>EDA was coined, developed and advocated for by John Tukey. His book, entitled
<em>“Exploratory Data Analysis”</em>
was published in 1977, and is still in use today. It may seem like an oddity, but it was a fundamental change in how data science / statistics was done. Fundamentally he sums up EDA with this quote:</p>
<p>
<em>“It is important to understand what you CAN DO, before you learn to measure how WELL you seem to have DONE it.”</em>
<em>– J. W. Tukey (1977)</em>
</p>
<p>If you don’t understand the data, it becomes difficult to know how to analyze it. Confirmatory and exploratory analyses are not superior or inferior to one another, rather they are complementary. With all the tools available to do both, ignoring
one of them is inexcusable.</p>
<p>
<em>“Today, exploratory and confirmatory (analysis) – can – and should – proceed side by side.”</em>
<em>– J. W. Tukey (1977)</em>
</p>
<h2>Cognitive Disfluency – make it work for you?</h2>
<p>There is often an urge due to productivity, laziness, or other factors to plow through with an analysis, using sophisticated analysis techniques to find the results you are looking for. With the proliferation of large datasets, this can be quite
ineffective, as it largely separates the analyst from the data, resulting in misunderstanding or not understanding the data at all.</p>
<p>There is some evidence that
<em>cognitive disfluency</em>
(making it harder to learn) can lead to deeper learning. For analysts and data scientists this means slowing things down, often using basic (and sometimes tedious) methods to integrate the primary structure and relationships contained in the data,
before pulling out the heavy machinery of modern data analysis.</p>
<p>See:
<em>Alter, A.L., 2013. The benefits of cognitive disfluency. Current Directions in Psychological Science, 22(6), pp.437-442.</em>
</p>
<p>All too often the success/failure of an analysis is determined from a single number, when in reality, understanding the data should be the goal.</p>
<p>When working with a new dataset I (Jesse) almost always start with what is discussed below. While cognitive disfluency deals with difficulty to learn, I believe slowing the analysis down, doing it carefully and critically leads to higher quality
analysis.</p>
</body>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@da1f583280eb43b19d8af3679ca36514" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">How to Begin</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@b1d57fce67c14554bd2f39f8bcbed787">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@b1d57fce67c14554bd2f39f8bcbed787" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p style="text-rendering: optimizelegibility; margin-top: 0px; margin-right: 0px; margin-left: 0px; padding: 0px; border: 0px; outline: 0px; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-stretch: inherit; font-family: 'Open Sans', 'Helvetica Neue', Helvetica, Arial, sans-serif; font-size: 18px; vertical-align: baseline; color: #313131;"><strong><span color="#313131" face="Open Sans, Helvetica Neue, Helvetica, Arial, sans-serif" style="color: #313131; font-family: 'Open Sans', 'Helvetica Neue', Helvetica, Arial, sans-serif;"><span style="font-size: 18px;">Importing a local file to RStudio</span></span></strong></p>
<p style="text-rendering: optimizelegibility; margin-top: 0px; margin-right: 0px; margin-left: 0px; padding: 0px; border: 0px; outline: 0px; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-stretch: inherit; font-family: 'Open Sans', 'Helvetica Neue', Helvetica, Arial, sans-serif; font-size: 18px; vertical-align: baseline; color: #313131;"><span style="color: #333333; font-size: 16px;">You can download the aline dataset</span><span style="color: #333333; font-size: 16px;"> to your computer <a href="/assets/courseware/v1/1d90e5ea3262695b6a1c14a377d178d8/asset-v1:MITx+HST.953x+3T2020+type@asset+block/aline_full_cohort_data.csv" target="[object Object]">here</a>.</span></p>
<p>You can then locate a data file on your computer and read it RStudio.</p>
<pre class="example" style="text-rendering: optimizelegibility; font-family: monaco, menlo, consolas, 'courier new', monospace; font-size: 14.4px; margin-top: 0px; padding: 9.5px; border: 1px solid #cccccc; outline: 0px; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-stretch: inherit; line-height: 1.42857; vertical-align: baseline; color: #333333; box-sizing: border-box; overflow: auto; border-radius: 4px; word-break: break-word; background-color: #f5f5f5;"># locate the file and set the working directory to the location where the file is stored<br />setwd("path_to_file")
dat <- read.csv("file_name.csv")</pre>
<p style="text-rendering: optimizelegibility; margin-top: 0px; margin-right: 0px; margin-left: 0px; padding: 0px; border: 0px; outline: 0px; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-stretch: inherit; font-family: 'Open Sans', 'Helvetica Neue', Helvetica, Arial, sans-serif; font-size: 18px; vertical-align: baseline; color: #313131;"></p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@88cce0f16ea84135ac22105d7b0b4504" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Summary Function in R</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@44a2cf3bc7ea430596bd0fc19c11706a">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@44a2cf3bc7ea430596bd0fc19c11706a" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>One form of EDA is to provide numerical summaries of the dataset. This can have many purposes:</p>
<ul>
<li>To verify the dataset you loaded is the one you think you did.</li>
<li>To quantify characteristics of the dataset which need to be reported numerically.</li>
</ul>
<h2>
The
<code>summary</code>
function in
<code>R</code>
</h2>
<p>
<code>R</code>
has a very handy function, which performs differently for depending on the type of data structures you apply it to. This is the
<code>summary</code>
function, and it provides a very useful data summary of data frames. This comes in the form of five-number summaries (plus the mean) (min-Q1-mean-median-Q3-max) for numeric data and counts for categorical (factor) data. Summary also works on many
types of objects in
<code>R</code>, and when you don’t know what to do with an
<code>R</code>
object (<code>obj</code>), it is often good to try
<code>summary(obj)</code>.
</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>summary(aline_full_cohort_data)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>aline_flg icu_los_day hospital_los_day age gender_num weight_first
Min. :0.0000 Min. : 0.500 Min. : 1.000 Min. :15.18 Min. :0.0000 Min. : 30.00
1st Qu.:0.0000 1st Qu.: 1.370 1st Qu.: 3.000 1st Qu.:38.25 1st Qu.:0.0000 1st Qu.: 65.40
Median :1.0000 Median : 2.185 Median : 6.000 Median :53.68 Median :1.0000 Median : 77.00
Mean :0.5541 Mean : 3.346 Mean : 8.111 Mean :54.38 Mean :0.5775 Mean : 80.08
3rd Qu.:1.0000 3rd Qu.: 4.003 3rd Qu.: 10.000 3rd Qu.:72.76 3rd Qu.:1.0000 3rd Qu.: 90.00
Max. :1.0000 Max. :28.240 Max. :112.000 Max. :99.11 Max. :1.0000 Max. :257.60
NA's :1 NA's :110
...
</code></pre>
<p>As you can see, this function is very verbose, but produces some useful output (only a part of the original output is shown here). At this point, it’s also a good idea to verify the number of rows and columns are correct:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>nrow(aline_full_cohort_data)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>## [1] 1776</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>ncol(aline_full_cohort_data)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>## [1] 46</code></pre>
<p>This should be what you’re expecting (<code>1776</code> and <code>46</code>). If it’s not, this could indicate a loading error, or a problem with the data extraction.</p>
<p>As you will note, many of the <code>flg</code> variables listed in the summary output above, are constrained by 0 and 1. This is because they have a binary encoding (usually 1 if present, and 0 if not). Although not necessary in this particular instance, it it sometimes useful to encode these types of variables as factors.
<br>
<br>The function <code>convert.bin.fac</code> will do this, and we’ll use it to create a new data frame called <code>dat2</code>, and call the summary function on the new data frame. We first need to load/install two packages (<code>devtools</code> and <code>mimicbook</code>).</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;">
<code>if(!("devtools" %in% installed.packages()[,1])) {
install.packages("devtools",repos="https://cloud.r-project.org")
}
library(devtools)
if(!("MIMICbook" %in% installed.packages()[,1])) {
install_github("jraffa/MIMICbook")
}
library(MIMICbook)</code></pre>
<p>The <code>MIMICbook</code> package provides some useful functions written for the textbook that we will use throughout some of the workshops for hst.953. It installs via github.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>dat2 <- convert.bin.fac(aline_full_cohort_data)
summary(dat2)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>aline_flg icu_los_day hospital_los_day age gender_num weight_first
0:792 Min. : 0.500 Min. : 1.000 Min. :15.18 Min. :0.0000 Min. : 30.00
1:984 1st Qu.: 1.370 1st Qu.: 3.000 1st Qu.:38.25 1st Qu.:0.0000 1st Qu.: 65.40
Median : 2.185 Median : 6.000 Median :53.68 Median :1.0000 Median : 77.00
Mean : 3.346 Mean : 8.111 Mean :54.38 Mean :0.5775 Mean : 80.08
3rd Qu.: 4.003 3rd Qu.: 10.000 3rd Qu.:72.76 3rd Qu.:1.0000 3rd Qu.: 90.00
Max. :28.240 Max. :112.000 Max. :99.11 Max. :1.0000 Max. :257.60
NA's :1 NA's :110
...
</code></pre>
<p>As you can now see (aline_flg), instead of means (which under the old encoding equate to proportions of patients where the variable == 1), now we have counts of patients with each <em>level</em> of the variable. This is because <code>R</code>’s summary function treats factors and numerical values differently.</p>
<p>Often, you will want to report these summaries separately for different groups. For instance, is the mean or median age the same for those who received aline, and those who didn’t? A multi-purpose function called <code>tapply</code> can help us with this.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;">
<code>tapply(dat2$age,dat2$aline_flg,summary)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>$`0`
Min. 1st Qu. Median Mean 3rd Qu. Max.
15.18 34.80 50.85 53.02 72.11 97.46
$`1`
Min. 1st Qu. Median Mean 3rd Qu. Max.
15.19 40.36 56.02 55.48 73.21 99.11 </code></pre>
<p>
This function stratifies the first argument (<code>age</code>) by the second argument (<code>aline_flg</code>) and run the third argument (<code>summary</code>) on it. So, in our case, run the <code>summary</code>
function on <code>age</code> for those who received aline (<code>aline_flg</code> = 1) and those who didn’t (<code>aline_flg</code> = 0).
</p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@fab8f69145b54afc91f9d675e45b572f" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Questions</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@2c90baed78c04187ad8ca8f8a832c410">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@2c90baed78c04187ad8ca8f8a832c410" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p style="text-rendering: optimizelegibility; margin: 0px 0px 1.41575em; padding: 0px; border: 0px; outline: 0px; font-variant-numeric: inherit; font-variant-east-asian: inherit; font-stretch: inherit; font-size: 18px; line-height: inherit; font-family: 'Open Sans', 'Helvetica Neue', Helvetica, Arial, sans-serif; vertical-align: baseline; color: #333333;">Using the <strong><span style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-style: inherit; font-variant: inherit; font-stretch: inherit; font-size: inherit; line-height: inherit; font-family: inherit; vertical-align: baseline;"><em style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-variant: inherit; font-weight: inherit; font-stretch: inherit; font-size: inherit; line-height: inherit; font-family: inherit; vertical-align: baseline;">dat2</em></span></strong> data frame, run the summary function for <em style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-variant: inherit; font-weight: inherit; font-stretch: inherit; font-size: inherit; line-height: inherit; font-family: inherit; vertical-align: baseline;">sofa_first</em> for those patients with an arterial catheter and those without (<em>aline_flg</em>). Then, run the summary function for <em style="text-rendering: optimizelegibility; margin: 0px; padding: 0px; border: 0px; outline: 0px; font-variant: inherit; font-weight: inherit; font-stretch: inherit; font-size: inherit; line-height: inherit; font-family: inherit; vertical-align: baseline;">sofa_first </em>for those patients who died within 28 days, and those who survived (<strong><em>day_28_flg</em></strong>).</p>
</div>
</div>
<div class="vert vert-1" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@38aeecef02d345d3a76d714fde3107ec">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@38aeecef02d345d3a76d714fde3107ec" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_38aeecef02d345d3a76d714fde3107ec" class="problems-wrapper" role="group"
aria-labelledby="38aeecef02d345d3a76d714fde3107ec-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@38aeecef02d345d3a76d714fde3107ec" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@38aeecef02d345d3a76d714fde3107ec/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="38aeecef02d345d3a76d714fde3107ec-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@38aeecef02d345d3a76d714fde3107ec-problem-progress" tabindex="-1">
Question I
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@38aeecef02d345d3a76d714fde3107ec-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>Based on the first calculated SOFA score, do patients who received an arterial catheter differ from patients who did not have an arterial line?</p>
<div class="choicegroup capa_inputtype" id="inputtype_38aeecef02d345d3a76d714fde3107ec_2_1">
<fieldset aria-describedby="status_38aeecef02d345d3a76d714fde3107ec_2_1">
<div class="field">
<input type="radio" name="input_38aeecef02d345d3a76d714fde3107ec_2_1" id="input_38aeecef02d345d3a76d714fde3107ec_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="38aeecef02d345d3a76d714fde3107ec_2_1-choice_0-label" for="input_38aeecef02d345d3a76d714fde3107ec_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_38aeecef02d345d3a76d714fde3107ec_2_1"> Yes, patients with an arterial line tend to be sicker and have a higher SOFA score on admission
</label>
</div>
<div class="field">
<input type="radio" name="input_38aeecef02d345d3a76d714fde3107ec_2_1" id="input_38aeecef02d345d3a76d714fde3107ec_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="38aeecef02d345d3a76d714fde3107ec_2_1-choice_1-label" for="input_38aeecef02d345d3a76d714fde3107ec_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_38aeecef02d345d3a76d714fde3107ec_2_1"> No, there is almost no difference in first SOFA scores
</label>
</div>
<div class="field">
<input type="radio" name="input_38aeecef02d345d3a76d714fde3107ec_2_1" id="input_38aeecef02d345d3a76d714fde3107ec_2_1_choice_2" class="field-input input-radio" value="choice_2"/><label id="38aeecef02d345d3a76d714fde3107ec_2_1-choice_2-label" for="input_38aeecef02d345d3a76d714fde3107ec_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_38aeecef02d345d3a76d714fde3107ec_2_1"> We can't say and need a fancier statistical model to answer this question
</label>
</div>
<span id="answer_38aeecef02d345d3a76d714fde3107ec_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_38aeecef02d345d3a76d714fde3107ec_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><div class="solution-span">
<span id="solution_38aeecef02d345d3a76d714fde3107ec_solution_1"/>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question I" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_38aeecef02d345d3a76d714fde3107ec" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_38aeecef02d345d3a76d714fde3107ec">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="38aeecef02d345d3a76d714fde3107ec-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="38aeecef02d345d3a76d714fde3107ec-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="38aeecef02d345d3a76d714fde3107ec-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
<div class="vert vert-2" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@36becf0e9246438597e59931e387f48b">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@36becf0e9246438597e59931e387f48b" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_36becf0e9246438597e59931e387f48b" class="problems-wrapper" role="group"
aria-labelledby="36becf0e9246438597e59931e387f48b-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@36becf0e9246438597e59931e387f48b" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@36becf0e9246438597e59931e387f48b/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="36becf0e9246438597e59931e387f48b-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@36becf0e9246438597e59931e387f48b-problem-progress" tabindex="-1">
Question II
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@36becf0e9246438597e59931e387f48b-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>Based on your analysis of the dataset, select all the <strong>WRONG</strong> answers.</p>
<div class="choicegroup capa_inputtype" id="inputtype_36becf0e9246438597e59931e387f48b_2_1">
<fieldset aria-describedby="status_36becf0e9246438597e59931e387f48b_2_1">
<div class="field">
<input type="checkbox" name="input_36becf0e9246438597e59931e387f48b_2_1[]" id="input_36becf0e9246438597e59931e387f48b_2_1_choice_0" class="field-input input-checkbox" value="choice_0"/><label id="36becf0e9246438597e59931e387f48b_2_1-choice_0-label" for="input_36becf0e9246438597e59931e387f48b_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_36becf0e9246438597e59931e387f48b_2_1"> Patients who have a lower SOFA score at admission frequently get invasive monitoring
</label>
</div>
<div class="field">
<input type="checkbox" name="input_36becf0e9246438597e59931e387f48b_2_1[]" id="input_36becf0e9246438597e59931e387f48b_2_1_choice_1" class="field-input input-checkbox" value="choice_1"/><label id="36becf0e9246438597e59931e387f48b_2_1-choice_1-label" for="input_36becf0e9246438597e59931e387f48b_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_36becf0e9246438597e59931e387f48b_2_1"> It is more common to place an arterial line in older and sicker patients
</label>
</div>
<div class="field">
<input type="checkbox" name="input_36becf0e9246438597e59931e387f48b_2_1[]" id="input_36becf0e9246438597e59931e387f48b_2_1_choice_2" class="field-input input-checkbox" value="choice_2"/><label id="36becf0e9246438597e59931e387f48b_2_1-choice_2-label" for="input_36becf0e9246438597e59931e387f48b_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_36becf0e9246438597e59931e387f48b_2_1"> Patients who tend to be younger but have a lower SOFA score get an arterial line placement more often
</label>
</div>
<span id="answer_36becf0e9246438597e59931e387f48b_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_36becf0e9246438597e59931e387f48b_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><div class="solution-span">
<span id="solution_36becf0e9246438597e59931e387f48b_solution_1"/>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question II" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_36becf0e9246438597e59931e387f48b" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_36becf0e9246438597e59931e387f48b">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="36becf0e9246438597e59931e387f48b-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="36becf0e9246438597e59931e387f48b-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="36becf0e9246438597e59931e387f48b-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
<div class="vert vert-3" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@fe06c8af22b84b67855bfb97347f1f9a">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@fe06c8af22b84b67855bfb97347f1f9a" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_fe06c8af22b84b67855bfb97347f1f9a" class="problems-wrapper" role="group"
aria-labelledby="fe06c8af22b84b67855bfb97347f1f9a-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@fe06c8af22b84b67855bfb97347f1f9a" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@fe06c8af22b84b67855bfb97347f1f9a/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="fe06c8af22b84b67855bfb97347f1f9a-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@fe06c8af22b84b67855bfb97347f1f9a-problem-progress" tabindex="-1">
Question III
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@fe06c8af22b84b67855bfb97347f1f9a-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>What is the median SOFA score in patients who died after 28 days in the ICU?</p>
<div class="choicegroup capa_inputtype" id="inputtype_fe06c8af22b84b67855bfb97347f1f9a_2_1">
<fieldset aria-describedby="status_fe06c8af22b84b67855bfb97347f1f9a_2_1">
<div class="field">
<input type="radio" name="input_fe06c8af22b84b67855bfb97347f1f9a_2_1" id="input_fe06c8af22b84b67855bfb97347f1f9a_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="fe06c8af22b84b67855bfb97347f1f9a_2_1-choice_0-label" for="input_fe06c8af22b84b67855bfb97347f1f9a_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_fe06c8af22b84b67855bfb97347f1f9a_2_1"> 0
</label>
</div>
<div class="field">
<input type="radio" name="input_fe06c8af22b84b67855bfb97347f1f9a_2_1" id="input_fe06c8af22b84b67855bfb97347f1f9a_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="fe06c8af22b84b67855bfb97347f1f9a_2_1-choice_1-label" for="input_fe06c8af22b84b67855bfb97347f1f9a_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_fe06c8af22b84b67855bfb97347f1f9a_2_1"> 6
</label>
</div>
<div class="field">
<input type="radio" name="input_fe06c8af22b84b67855bfb97347f1f9a_2_1" id="input_fe06c8af22b84b67855bfb97347f1f9a_2_1_choice_2" class="field-input input-radio" value="choice_2"/><label id="fe06c8af22b84b67855bfb97347f1f9a_2_1-choice_2-label" for="input_fe06c8af22b84b67855bfb97347f1f9a_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_fe06c8af22b84b67855bfb97347f1f9a_2_1"> 3.043
</label>
</div>
<span id="answer_fe06c8af22b84b67855bfb97347f1f9a_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_fe06c8af22b84b67855bfb97347f1f9a_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><div class="solution-span">
<span id="solution_fe06c8af22b84b67855bfb97347f1f9a_solution_1"/>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question III" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_fe06c8af22b84b67855bfb97347f1f9a" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_fe06c8af22b84b67855bfb97347f1f9a">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="fe06c8af22b84b67855bfb97347f1f9a-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="fe06c8af22b84b67855bfb97347f1f9a-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="fe06c8af22b84b67855bfb97347f1f9a-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@3dec87969ad64858ab42a8d68cbe9949" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Tables</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@bbb4d7c1d69d49758eaa1f6305001277">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@bbb4d7c1d69d49758eaa1f6305001277" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>The output from summary is very useful, but is generally not acceptable in formal research reports, let alone a published paper. There are several ways to produce a publication which has a better layout. One way is described in the text book (Chapter 15), another, which we will cover here is through an <code>R</code> package called <code>tableone</code>.</p>
<p>As some of you may know, “Table 1” often refers to the table presented in most medical manuscripts which contains information used to describe the cohort. This often includes information about average patient age, gender distribution, and other important demographic, clinical and socioeconomic characteristics. We will cover briefly how to use the <code>CreateTableOne</code> function in this package to generate a table which is closer to being publication worthy.</p>
<p>The following code will install the <code>tableone</code> package and load it.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>if(!("tableone" %in% installed.packages()[,1])) {
install.packages("tableone",repos="https://cloud.r-project.org")
}
library(tableone)</code></pre>
<p>Here is an example functional call to <code>CreateTableOne</code>, which computes either the mean and standard deviation for numeric variables, or count and percentage for factors. You specify which variables you want to include in the table</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>CreateTableOne(vars=c("age","service_unit","aline_flg","day_28_flg"),
data=dat2)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code> Overall
n 1776
age (mean (SD)) 54.38 (21.06)
service_unit (%)
FICU 62 ( 3.5)
MICU 732 (41.2)
SICU 982 (55.3)
aline_flg = 1 (%) 984 (55.4)
day_28_flg = 1 (%) 283 (15.9)
</code></pre>
<p>We may want to breakdown these summaries further, like we did above with <code>tapply</code>, but we can do it with one function with the <code>CreateTableOne</code> function by passing the <code>strata</code> parameter. <code>strata</code> specifies which variable to stratify (breakdown) the others by. For example, here is the same table in the previous chunk, broken down by whether a patient received an aline or not.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>CreateTableOne(vars=c("age","service_unit","aline_flg","day_28_flg"),
strata="aline_flg",
data=dat2,
test=FALSE)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code> Stratified by aline_flg
0 1
n 792 984
age (mean (SD)) 53.02 (21.67) 55.48 (20.51)
service_unit (%)
FICU 24 ( 3.0) 38 ( 3.9)
MICU 480 (60.6) 252 ( 25.6)
SICU 288 (36.4) 694 ( 70.5)
aline_flg = 1 (%) 0 ( 0.0) 984 (100.0)
day_28_flg = 1 (%) 113 (14.3) 170 ( 17.3) </code></pre>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@693af97b3ebf442bbe3582a07310e1e6" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Questions</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@7148f95cc3b54d228f58ae8539a5db52">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@7148f95cc3b54d228f58ae8539a5db52" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>Compute a Table to summarise those variables considered before (<em>age, service_unit, aline_flg </em>and<em> day_28_flg</em>) , but now stratify by survival at 28 days (<em><strong>day_28_flg</strong></em>)</p>
</div>
</div>
<div class="vert vert-1" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@ec542a46332a448fbaebcca11188b3d5">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@ec542a46332a448fbaebcca11188b3d5" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_ec542a46332a448fbaebcca11188b3d5" class="problems-wrapper" role="group"
aria-labelledby="ec542a46332a448fbaebcca11188b3d5-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@ec542a46332a448fbaebcca11188b3d5" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@ec542a46332a448fbaebcca11188b3d5/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="ec542a46332a448fbaebcca11188b3d5-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@ec542a46332a448fbaebcca11188b3d5-problem-progress" tabindex="-1">
Question I
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@ec542a46332a448fbaebcca11188b3d5-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>Choose the <strong>incorrect</strong> answer.</p>
<div class="choicegroup capa_inputtype" id="inputtype_ec542a46332a448fbaebcca11188b3d5_2_1">
<fieldset aria-describedby="status_ec542a46332a448fbaebcca11188b3d5_2_1">
<div class="field">
<input type="radio" name="input_ec542a46332a448fbaebcca11188b3d5_2_1" id="input_ec542a46332a448fbaebcca11188b3d5_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="ec542a46332a448fbaebcca11188b3d5_2_1-choice_0-label" for="input_ec542a46332a448fbaebcca11188b3d5_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_ec542a46332a448fbaebcca11188b3d5_2_1"> 283 patients had died in the ICU after 28 days of their stay
</label>
</div>
<div class="field">
<input type="radio" name="input_ec542a46332a448fbaebcca11188b3d5_2_1" id="input_ec542a46332a448fbaebcca11188b3d5_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="ec542a46332a448fbaebcca11188b3d5_2_1-choice_1-label" for="input_ec542a46332a448fbaebcca11188b3d5_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_ec542a46332a448fbaebcca11188b3d5_2_1"> 113 patients who died in the ICU after 28 days (<em>day_28_flg = 1</em>) had no arterial line.
</label>
</div>
<div class="field">
<input type="radio" name="input_ec542a46332a448fbaebcca11188b3d5_2_1" id="input_ec542a46332a448fbaebcca11188b3d5_2_1_choice_2" class="field-input input-radio" value="choice_2"/><label id="ec542a46332a448fbaebcca11188b3d5_2_1-choice_2-label" for="input_ec542a46332a448fbaebcca11188b3d5_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_ec542a46332a448fbaebcca11188b3d5_2_1"> 170 patients who where alife after 28 days have recieved an arterial line
</label>
</div>
<span id="answer_ec542a46332a448fbaebcca11188b3d5_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_ec542a46332a448fbaebcca11188b3d5_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><div class="solution-span">
<span id="solution_ec542a46332a448fbaebcca11188b3d5_solution_1"/>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question I" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_ec542a46332a448fbaebcca11188b3d5" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_ec542a46332a448fbaebcca11188b3d5">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
<span class="problem-action-button-wrapper">
<button type="button" class="show problem-action-btn btn-default btn-small" aria-describedby="ec542a46332a448fbaebcca11188b3d5-problem-title"><span class="icon fa fa-info-circle" aria-hidden="true"></span><span class="show-label">Show Answer</span></button>
</span>
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="ec542a46332a448fbaebcca11188b3d5-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="ec542a46332a448fbaebcca11188b3d5-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="ec542a46332a448fbaebcca11188b3d5-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
<div class="vert vert-2" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@084fd35dff6a4872a13b9fa4eee44296">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@084fd35dff6a4872a13b9fa4eee44296" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_084fd35dff6a4872a13b9fa4eee44296" class="problems-wrapper" role="group"
aria-labelledby="084fd35dff6a4872a13b9fa4eee44296-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@084fd35dff6a4872a13b9fa4eee44296" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@084fd35dff6a4872a13b9fa4eee44296/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="084fd35dff6a4872a13b9fa4eee44296-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@084fd35dff6a4872a13b9fa4eee44296-problem-progress" tabindex="-1">
Question II
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@084fd35dff6a4872a13b9fa4eee44296-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>Choose one incorrect expression that is not helpful for reproducing the mean age statistic stratified by <em>aline_flg</em>.</p>
<div class="choicegroup capa_inputtype" id="inputtype_084fd35dff6a4872a13b9fa4eee44296_2_1">
<fieldset aria-describedby="status_084fd35dff6a4872a13b9fa4eee44296_2_1">
<div class="field">
<input type="radio" name="input_084fd35dff6a4872a13b9fa4eee44296_2_1" id="input_084fd35dff6a4872a13b9fa4eee44296_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="084fd35dff6a4872a13b9fa4eee44296_2_1-choice_0-label" for="input_084fd35dff6a4872a13b9fa4eee44296_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_084fd35dff6a4872a13b9fa4eee44296_2_1"> tapply(dat2$age,dat2$aline_flg,mean)
</label>
</div>
<div class="field">
<input type="radio" name="input_084fd35dff6a4872a13b9fa4eee44296_2_1" id="input_084fd35dff6a4872a13b9fa4eee44296_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="084fd35dff6a4872a13b9fa4eee44296_2_1-choice_1-label" for="input_084fd35dff6a4872a13b9fa4eee44296_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_084fd35dff6a4872a13b9fa4eee44296_2_1"> mean(dat2$age[dat2$aline_flg==0])
</label>
</div>
<div class="field">
<input type="radio" name="input_084fd35dff6a4872a13b9fa4eee44296_2_1" id="input_084fd35dff6a4872a13b9fa4eee44296_2_1_choice_2" class="field-input input-radio" value="choice_2"/><label id="084fd35dff6a4872a13b9fa4eee44296_2_1-choice_2-label" for="input_084fd35dff6a4872a13b9fa4eee44296_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_084fd35dff6a4872a13b9fa4eee44296_2_1"> tapply(dat2$age,dat2$aline_flg,summary)
</label>
</div>
<span id="answer_084fd35dff6a4872a13b9fa4eee44296_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_084fd35dff6a4872a13b9fa4eee44296_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><div class="solution-span">
<span id="solution_084fd35dff6a4872a13b9fa4eee44296_solution_1"/>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question II" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_084fd35dff6a4872a13b9fa4eee44296" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_084fd35dff6a4872a13b9fa4eee44296">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
<span class="problem-action-button-wrapper">
<button type="button" class="show problem-action-btn btn-default btn-small" aria-describedby="084fd35dff6a4872a13b9fa4eee44296-problem-title"><span class="icon fa fa-info-circle" aria-hidden="true"></span><span class="show-label">Show Answer</span></button>
</span>
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="084fd35dff6a4872a13b9fa4eee44296-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="084fd35dff6a4872a13b9fa4eee44296-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="084fd35dff6a4872a13b9fa4eee44296-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@7c998012ccb949fc8ca7dd92c8671d05" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Optional</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@dda176f242134d31baffa7172bfdee75">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@dda176f242134d31baffa7172bfdee75" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>As an aside, the following code may help for your projects, as it improves the presentation of the tables above. You will still need to update the column and row names manually, but this should paste nicely into Word or LateX!</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>if(!("dplyr" %in% installed.packages()[,1])) {
install.packages("dplyr")
}
library(dplyr)
CreateTableOne(vars=c("age","service_unit","aline_flg","day_28_flg"),
strata="aline_flg",
data=dat2,
test=FALSE) %>%
print(printToggle=FALSE,
showAllLevels=TRUE,
cramVars="kon") %>%
{data.frame(
variable_name= gsub(" ", "&nbsp;", rownames(.), fixed = TRUE), .,
row.names=NULL,
check.names=FALSE,
stringsAsFactors=FALSE)} %>%
knitr::kable()</code></pre>
<table>
<thead>
<tr class="header">
<th align="left">variable_name</th>
<th align="left">level</th>
<th align="left">0</th>
<th align="left">1</th>
</tr>
</thead>
<tbody>
<tr class="odd">
<td align="left">n</td>
<td align="left"></td>
<td align="left">792</td>
<td align="left">984</td>
</tr>
<tr class="even">
<td align="left">age (mean (sd))</td>
<td align="left"></td>
<td align="left">753.02 (21.67)</td>
<td align="left">55.48 (20.51)</td>
</tr>
<tr class="odd">
<td align="left">service_unit (%)</td>
<td align="left">FICU</td>
<td align="left">24 ( 3.0)</td>
<td align="left">38 ( 3.9)</td>
</tr>
<tr class="even">
<td align="left"></td>
<td align="left">MICU</td>
<td align="left">480 ( 60.6)</td>
<td align="left">252 ( 25.6)</td>
</tr>
<tr class="odd">
<td align="left"></td>
<td align="left">SICU</td>
<td align="left">288 ( 36.4) </td>
<td align="left">694 ( 70.5)</td>
</tr>
<tr class="odd">
<td align="left">aline_flg (%)</td>
<td align="left">0</td>
<td align="left">792 (100.0)</td>
<td align="left">0 ( 0.0)</td>
</tr>
<tr class="even">
<td align="left"></td>
<td align="left">1</td>
<td align="left">0 ( 0.0)</td>
<td align="left">984 (100.0)</td>
</tr>
<tr class="odd">
<td align="left">day_28_flg (%)</td>
<td align="left">0</td>
<td align="left">679 ( 85.7) </td>
<td align="left">814 ( 82.7)</td>
</tr>
<tr class="even">
<td align="left"></td>
<td align="left">1</td>
<td align="left">113 ( 14.3)</td>
<td align="left">170 ( 17.3) </td>
</tr>
</tbody>
</table>
</div>
<script>
// add bootstrap table styles to pandoc tables
function bootstrapStylePandocTables() {
$('tr.header').parent('thead').parent('table').addClass('table table-condensed');
}
$(document).ready(function () {
bootstrapStylePandocTables();
});
</script>
</body>
</html>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@15117cf85cb543729dde400a55bc7062" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Other Bivariate Numerical Summaries</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@3ecae8799afc4fb7bc97912e34895c32">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@3ecae8799afc4fb7bc97912e34895c32" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>Sometimes you may wish to display the relationships between two or more variables directly. For categorical variables this can be tricky. One common way to explore relationships between categorical variables is by producing the cross tabulated tables (“crosstabs” for short). This is mainly done via the <code>table</code> function, which can take several categorical variables, and produce the number of patients which meet criteria for those variables. For instance, looking at how aline was used in men and women:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>table(dat2$gender_num,dat2$aline_flg,dnn=c("Gender","Aline"))</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code> Aline
Gender 0 1
0 344 406
1 447 578</code></pre>
<p>We can see that aline was used 578 times in men (<code>gender_num=1</code>) and 406 times in women (<code>gender_num=0</code>). The raw numbers are often difficult to compare, so often the proportions are more useful. Applying <code>prop.table</code> to our existing table, and adding the argument 1 (for by row, use 2 for columns), we get the proportion of men and women who had aline (56% vs. 54%).</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>prop.table(table(dat2$gender_num,dat2$aline_flg,dnn=c("Gender","Aline")),1)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code> Aline
Gender 0 1
0 0.4586667 0.5413333
1 0.4360976 0.5639024</code></pre>
<p>A different summary for the relationships exists when dealing with bivariate numeric data. Sometimes it’s desirable to present the strength of the relationship between two variables, and correlation coefficient is the way to go about this. There is a <code>cor</code> function in <code>R</code>, but when dealing with only two variables, it’s easiest to use the <code>cor.test</code> function. Under the defaults, it computes the Pearson product-moment correlation and computes a hypothesis test to assess if there’s evidence that the correlation is not zero. Alternative forms of correlation are computed below, including Spearman’s rho and Kendall’s tau. These latter methods are useful when dealing with data which is not necessarily numeric but ordered (e.g., likert based rankings on a 1-5 scale) or has outliers. Spearman’s rho and Kendall’s tau are rank based methods, and also have a certain degree of robustness to outliers in the data. None of these methods are robust to non-linear relationships, and it’s very easy to miss a strong relationship between two variables if you rely on these methods in isolation.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>cor.test(dat2$bun_first,dat2$creatinine_first)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>Pearson's product-moment correlation
data: dat2$bun_first and dat2$creatinine_first
t = 33.233, df = 1768, p-value < 2.2e-16
alternative hypothesis: true correlation is not equal to 0
95 percent confidence interval:
0.5905441 0.6479479
sample estimates:
cor
0.6200752 </code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>cor.test(dat2$bun_first,dat2$creatinine_first,method="spearman")</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code># Cannot compute exact p-value with ties
Spearman's rank correlation rho
data: dat2$bun_first and dat2$creatinine_first
S = 977030000, p-value < 2.2e-16
alternative hypothesis: true rho is not equal to 0
sample estimates:
rho
0.5754121</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>cor.test(dat2$bun_first,dat2$creatinine_first,method="kendall")</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>endall's rank correlation tau
data: dat2$bun_first and dat2$creatinine_first
z = 24.807, p-value < 2.2e-16
alternative hypothesis: true tau is not equal to 0
sample estimates:
tau
0.418814 </code></pre>
<p>We can produce a scatterplot of the same two variables:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>with(dat2,plot(bun_first,creatinine_first))</code></pre>
<p><img src="data:image/png;base64,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" width="960"/></p>
<p>We can see that there is indeed a positive correlation between the two variables, but the data has more variability for higher values of <code>bun_first</code> and <code>creatinine_first</code>. It’s advisable to consider transformations of these two variables and be wary about using Pearson’s correlation.</p>
<p>Going beyond two dimensions can be a little tricky. Plotting on a three dimensional axis, while possible, is not ideal, and I know very few people who can see in four dimensions.</p>
<p>What is possible, is to use other aspects of the plot (e.g., size, color, shape, hue, transparency, location) to identify features you would like to see. For instance, in the above plot, we can add color to identify those who died:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>with(dat2,plot(bun_first,creatinine_first,col=day_28_flg,pch=19))</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@3a33bcce8fb14edcb9f9d0b9e8019bca" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Questions</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@f7c0eb4e0caa462bbe8f0cc0e0ac7db0">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@f7c0eb4e0caa462bbe8f0cc0e0ac7db0" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_f7c0eb4e0caa462bbe8f0cc0e0ac7db0" class="problems-wrapper" role="group"
aria-labelledby="f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@f7c0eb4e0caa462bbe8f0cc0e0ac7db0" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@f7c0eb4e0caa462bbe8f0cc0e0ac7db0/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-progress" tabindex="-1">
Question I
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>Calculate the number of female patients who where alive after 28 days in the ICU.</p>
<div class="choicegroup capa_inputtype" id="inputtype_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1">
<fieldset aria-describedby="status_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1">
<div class="field">
<input type="radio" name="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1" id="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1-choice_0-label" for="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1"> 606
</label>
</div>
<div class="field">
<input type="radio" name="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1" id="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1-choice_1-label" for="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1"> 144
</label>
</div>
<div class="field">
<input type="radio" name="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1" id="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_2" class="field-input input-radio" value="choice_2"/><label id="f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1-choice_2-label" for="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1"> 886
</label>
</div>
<div class="field">
<input type="radio" name="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1" id="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_3" class="field-input input-radio" value="choice_3"/><label id="f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1-choice_3-label" for="input_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1_choice_3" class="response-label field-label label-inline" aria-describedby="status_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1"> 139
</label>
</div>
<span id="answer_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><div class="solution-span">
<span id="solution_f7c0eb4e0caa462bbe8f0cc0e0ac7db0_solution_1"/>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question I" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_f7c0eb4e0caa462bbe8f0cc0e0ac7db0" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_f7c0eb4e0caa462bbe8f0cc0e0ac7db0">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="f7c0eb4e0caa462bbe8f0cc0e0ac7db0-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
<div class="vert vert-1" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@edb47b04a40f4719b952ab3473defb46">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@edb47b04a40f4719b952ab3473defb46" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_edb47b04a40f4719b952ab3473defb46" class="problems-wrapper" role="group"
aria-labelledby="edb47b04a40f4719b952ab3473defb46-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@edb47b04a40f4719b952ab3473defb46" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@edb47b04a40f4719b952ab3473defb46/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="edb47b04a40f4719b952ab3473defb46-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@edb47b04a40f4719b952ab3473defb46-problem-progress" tabindex="-1">
Question II
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@edb47b04a40f4719b952ab3473defb46-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><p>Calculate the proportion of men who died after 28 day of the ICU stay</p>
<div class="choicegroup capa_inputtype" id="inputtype_edb47b04a40f4719b952ab3473defb46_2_1">
<fieldset aria-describedby="status_edb47b04a40f4719b952ab3473defb46_2_1">
<div class="field">
<input type="radio" name="input_edb47b04a40f4719b952ab3473defb46_2_1" id="input_edb47b04a40f4719b952ab3473defb46_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="edb47b04a40f4719b952ab3473defb46_2_1-choice_0-label" for="input_edb47b04a40f4719b952ab3473defb46_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_edb47b04a40f4719b952ab3473defb46_2_1"> 0.183
</label>
</div>
<div class="field">
<input type="radio" name="input_edb47b04a40f4719b952ab3473defb46_2_1" id="input_edb47b04a40f4719b952ab3473defb46_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="edb47b04a40f4719b952ab3473defb46_2_1-choice_1-label" for="input_edb47b04a40f4719b952ab3473defb46_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_edb47b04a40f4719b952ab3473defb46_2_1"> 0.491
</label>
</div>
<div class="field">
<input type="radio" name="input_edb47b04a40f4719b952ab3473defb46_2_1" id="input_edb47b04a40f4719b952ab3473defb46_2_1_choice_2" class="field-input input-radio" value="choice_2"/><label id="edb47b04a40f4719b952ab3473defb46_2_1-choice_2-label" for="input_edb47b04a40f4719b952ab3473defb46_2_1_choice_2" class="response-label field-label label-inline" aria-describedby="status_edb47b04a40f4719b952ab3473defb46_2_1"> 0.871
</label>
</div>
<span id="answer_edb47b04a40f4719b952ab3473defb46_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_edb47b04a40f4719b952ab3473defb46_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div><p>[explanation]</p>
<p>prop.table(table(dat2$gender_num[dat2$day_28_flg==1]))</p>
<p>[explanation]</p>
</div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question II" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_edb47b04a40f4719b952ab3473defb46" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_edb47b04a40f4719b952ab3473defb46">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
<span class="problem-action-button-wrapper">
<button type="button" class="show problem-action-btn btn-default btn-small" aria-describedby="edb47b04a40f4719b952ab3473defb46-problem-title"><span class="icon fa fa-info-circle" aria-hidden="true"></span><span class="show-label">Show Answer</span></button>
</span>
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="edb47b04a40f4719b952ab3473defb46-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="edb47b04a40f4719b952ab3473defb46-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="edb47b04a40f4719b952ab3473defb46-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@02c1123551b14ef2a0d39a0e85e3eadd" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Creating Categorical Variables From Continuous/Numeric Variables</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@5e923e42475b4f63913f162d820e9dd2">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@5e923e42475b4f63913f162d820e9dd2" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>Sometimes numeric variables need to be broken down into categorical variables or factors. This can be done for a variety of reasons. There is a useful function called <code>cut2</code> in the <code>Hmisc</code> package. We install it and use it below.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>if(!("Hmisc" %in% installed.packages()[,1])) {
install.packages("Hmisc",repos="https://cloud.r-project.org")
}
library(Hmisc)
dat2$age.cat <- cut2(dat2$age,g=5)
table(dat2$age.cat)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>[15.2,33.6) [33.6,47.6) [47.6,60.3) [60.3,76.4) [76.4,99.1]
356 355 355 355 355 </code></pre>
<p><code>cut2</code> typically needs two arguments. The first is a numeric variable to convert into a factor, and the second is how to do the splitting. Specifying <code>g=5</code> (as above for <code>age.cat</code>) breaks the numeric variable into 5 groups, with the cut points determined by attempting to make the groups as equally sized as possible. <br><br></p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>dat2$age.cat2 <- cut2(dat2$age,c(25,40,55,70,85))
table(dat2$age.cat2)</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>[15.2,25.0) [25.0,40.0) [40.0,55.0) [55.0,70.0) [70.0,85.0) [85.0,99.1]
208 287 428 341 382 130
</code></pre>
<p>As you can see in this example, due to the odd number of patients, they are not perfectly even. The second approach requires passing the cut points. In the second example, we tell <code>R</code> to cut the data at 25, 40, … . This results in 6 groups for five cut points.</p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@efb574d1fef24134b37d10f7b9632b15" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Questions</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@d731f736178349c3a5bc3efbdc49e28b">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@d731f736178349c3a5bc3efbdc49e28b" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>Create a new variable in the dat2 data frame called sofa.cat made up of four (aprox.) equally sized groups for SOFA. Print the sample size in each group. Why might the number in each group differ so much?</p>
<p></p>
</div>
</div>
<div class="vert vert-1" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@2c62fa41a1d842a19f1df25b3cc034f0">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@2c62fa41a1d842a19f1df25b3cc034f0" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_2c62fa41a1d842a19f1df25b3cc034f0" class="problems-wrapper" role="group"
aria-labelledby="2c62fa41a1d842a19f1df25b3cc034f0-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@2c62fa41a1d842a19f1df25b3cc034f0" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@2c62fa41a1d842a19f1df25b3cc034f0/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="1"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="2c62fa41a1d842a19f1df25b3cc034f0-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@2c62fa41a1d842a19f1df25b3cc034f0-problem-progress" tabindex="-1">
Question I
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@2c62fa41a1d842a19f1df25b3cc034f0-problem-progress"></div>
<div class="problem">
<div>
<div class="wrapper-problem-response" tabindex="-1" aria-label="Question 1" role="group"><div class="choicegroup capa_inputtype" id="inputtype_2c62fa41a1d842a19f1df25b3cc034f0_2_1">
<fieldset aria-describedby="status_2c62fa41a1d842a19f1df25b3cc034f0_2_1">
<legend id="2c62fa41a1d842a19f1df25b3cc034f0_2_1-legend" class="response-fieldset-legend field-group-hd">For each SOFA group calculate the number of people who survived and died in the hospital and at 28 days (use the hosp_exp_flg and day_28_flg variable). Does mortality go up or down as SOFA increases?</legend>
<div class="field">
<input type="radio" name="input_2c62fa41a1d842a19f1df25b3cc034f0_2_1" id="input_2c62fa41a1d842a19f1df25b3cc034f0_2_1_choice_0" class="field-input input-radio" value="choice_0"/><label id="2c62fa41a1d842a19f1df25b3cc034f0_2_1-choice_0-label" for="input_2c62fa41a1d842a19f1df25b3cc034f0_2_1_choice_0" class="response-label field-label label-inline" aria-describedby="status_2c62fa41a1d842a19f1df25b3cc034f0_2_1"> The mortality decreases
</label>
</div>
<div class="field">
<input type="radio" name="input_2c62fa41a1d842a19f1df25b3cc034f0_2_1" id="input_2c62fa41a1d842a19f1df25b3cc034f0_2_1_choice_1" class="field-input input-radio" value="choice_1"/><label id="2c62fa41a1d842a19f1df25b3cc034f0_2_1-choice_1-label" for="input_2c62fa41a1d842a19f1df25b3cc034f0_2_1_choice_1" class="response-label field-label label-inline" aria-describedby="status_2c62fa41a1d842a19f1df25b3cc034f0_2_1"> The mortality increases
</label>
</div>
<span id="answer_2c62fa41a1d842a19f1df25b3cc034f0_2_1"/>
</fieldset>
<div class="indicator-container">
<span class="status unanswered" id="status_2c62fa41a1d842a19f1df25b3cc034f0_2_1" data-tooltip="Not yet answered.">
<span class="sr">unanswered</span><span class="status-icon" aria-hidden="true"/>
</span>
</div>
</div></div>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Question I" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_2c62fa41a1d842a19f1df25b3cc034f0" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_2c62fa41a1d842a19f1df25b3cc034f0">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="2c62fa41a1d842a19f1df25b3cc034f0-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="2c62fa41a1d842a19f1df25b3cc034f0-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="2c62fa41a1d842a19f1df25b3cc034f0-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
<div class="vert vert-2" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@77fe59ed5e334b3b8d8b38805ae8ef3d">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@77fe59ed5e334b3b8d8b38805ae8ef3d" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>Can you suggest a reason why we might want to convert age to a categorical variable instead of using it in it’s current form?</p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@4c146390a9a94862a84838d5a39eebe4" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Plotting Relationships With Discrete Variables</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@33c6ec1c11e04def8676c0bdaf572ef1">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@33c6ec1c11e04def8676c0bdaf572ef1" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p>Plotting discrete data can be a little tricky, but if done right can be very effective. For an example of why it’s difficult, let’s plot two discrete variables: <code>gender_num</code> and <code>aline_flg</code>.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot(dat2$gender_num,dat2$aline_flg,xlab="Gender",ylab="IAC")</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
<p>Because we have converted <code>gender_num</code> and <code>aline_flg</code> to a factor, <code>R</code> gives us what is called a “Factor Plot”. The area of the light grey region is proportional to the proportion of each gender who received an aline. In this case, there is not that big of a difference between the genders.</p>
<p>This factor plot is more useful than if we were to keep the original numerical class both these variables had. For example, if we use the original <code>dat</code> data frame both <code>gender_num</code> and <code>aline_flg</code> are numeric variables, and when we plot these, we don’t end up with something very useful:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot(dat$gender_num,dat$aline_flg,xlab="Gender",ylab="IAC")</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
<p>In this case we only have four different types of data points, and although we could try to jitter the values to get a slightly more useful plot, it’s unlikely that this would give us a good visual interpretation of the data.</p>
<p>Sometimes the covariate may take on more than two levels. Here, we plot the in-hospital mortality rate by the different SOFA values, and put a smooth curve through the points. This covers a more <em>advanced</em> topic, and we <em><em>don’t</em></em> expect you to understand the technical details of the code below.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot(names(table(dat2$sofa_first)),
sapply(split(dat2,dat2$sofa_first),function(x) {mean(x$hosp_exp_flg==1,na.rm=T)}),
xlab="SOFA",
ylab="In-Hospital Mortality",
cex=log(as.numeric(table(dat2$sofa_first)+1))/5)
lines(smooth.spline(dat2$sofa_first,dat2$hosp_exp_flg==1),type="l")
# uncomment the following code if the smoth.spline functions returns an error
# "Error in smooth.spline(dat2$sofa_first, dat2$hosp_exp_flg == 1) : missing or infinite values in inputs are not allowed"
# lines(smooth.spline(dat2$sofa_first[!is.na(dat2$sofa_first)],
# dat2$hosp_exp_flg[!is.na(dat2$sofa_first)]==1),
# type="l")
</code></pre>
<p><img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAB4AAAAYACAMAAACgu6WYAAAA1VBMVEUAAAAAADoAAGYAOjoAOmYAOpAAZrY6AAA6OgA6Ojo6OmY6OpA6ZmY6ZpA6ZrY6kNtmAABmADpmOgBmOjpmZjpmZmZmZpBmkJBmkLZmkNtmtrZmtttmtv+QOgCQOjqQZgCQZjqQZmaQkDqQkGaQkLaQtraQttuQtv+Q2/+2ZgC2Zjq2kDq2kGa2kJC2kLa2tpC2tra2ttu227a229u22/+2///bkDrbkGbbtmbbtpDbtrbb25Db27bb29vb2//b////tmb/25D/27b/29v//7b//9v///8gpvFCAAAACXBIWXMAAB2HAAAdhwGP5fFlAAAgAElEQVR4nO3dfWPTWN6gabkaCDBFwaQ6NTwD7NA7T7GVGnYIC5UGhlDhxf7+H2kt5812TIgc6fzOObquP7qIm8g6tqwbyXppZgBAck30DADAGAkwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIkHeAGwDIQv+J632KPYp+tQHgTO+N63uCfRrgHxwAsAUBBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACFBqgKcfDl5/6v5rAgxAHkoK8Nejs+S+321ak1/+7jgJAQYgDwUF+Nuj5qe/2j9M/2zOTJ52m4YAA5CHEgO835b35729X9sEdyuwAAOQhwIDfDzP7sm+5y/PmtMmX5cAA5CHAgM83wC+c/rQdF7gh12mIcAA5KG8ALfRPd/vPN8avt3lQCwBBiAP5QX4/KvgpceuTYAByIMAA0CA8gI8228mf5w99nlHgAEoUWEBbst7vHTgle+AAShTWQGe+8d/e/M/zjeB24ccBQ1AgYoL8InFfufpv3ecBwxAmQoK8Ly4H1/+unMR4LbIF18HX4sAA5CHogK8sKjwWYBvve32ywIMQB7KC/CF6f/umF8BBiAXJQd4CwIMQB4EGAACCDAABBBgANjOjZpSeICvvhZ0s0HCmQOgajeLigADwFYEuMPfF2AA+jLqAM++Hn3q8tcFGIA8lB7gjgQYgDwIMAAEEGAACFBggKdHHw4ODl4f/b3F7wowAHkoLcAfny+dUvTgTddfF2AA8lBWgL/srp3Ve6vb7YAFGIBMFBXg4502uvf2Ttxtf5g87TQFAQYgDyUF+NujeXBfLD3wfh7kTtfhEGAAMlFSgA8v5bZN8sMukxBgAPJQUICnz5pmfYfzcdPc7nI0tAADkIeCArzpus+uBQ1AmQQYAAIUFODps2ayftaRXdAAlKmgAM/2L9W2/Vr4TpdJCDAAeSgpwJ935gV+u/TAl3l/L20UX0mAAchDSQFuz0OaF3fv94PWq5MrcXQ6C0mAAchEUQGevdtZuxTl5Em3CQgwAHkoK8Cz6cvlBE8ed70jkgADkIfCAjw3/Xjwcm9v7/HBmy3uRyjAAOShvADfiAADkAcBBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEgQLkBnn44eP13118SYADyUG6Avz1qfvqr6y8JMAB5EGAACFBSgL8eLfs4D/Cb+X8/dZmEAAOQh4ICPN/k3aTTZrAAA5AHAQaAAAUFePZup2kme2f+udNMfp7/97cuh0ILMAB5KCnAsy/PmubW29MfHIQFQMGKCvBs9u/5Zu9/nPxRgAEoWGEBnn3ZbZrbi41gAQagYKUFeDb9X00zeTITYACKVlyAZ7PP843gB38LMAAlKzDAs+mf843gFwIMQMFKDPDJCUk/7wgwAMUqM8CLE5I6XoNjQYAByEOhAV6ckCTAABSr2ADPvvyr20WwFgQYgDyUG+CtCDAAeRBgAAggwAAQoPAAX30y8Ka7FyacOQD4LgEGgABVB/gyAQYgD4UHePb16FOXvy7AAOSh9AB3JMAA5EGAASCAAANAgAIDPD36cHBw8Pqo83UoZwIMQC5KC/DH50unFD140/XXBRiAPJQV4C+7a2f13vqj2wQEGIA8FBXg4502uvf2Ttxtf5g87TQFAQYgDyUF+NujeXBfLD3wvvM9gQUYgDyUFODDS7ltk/ywyyQEGIA8FBTg6bOmWd/hfNw0t7scDS3AAOShoABvuu6za0EDUCYBBoAABQV4+qyZrJ91ZBc0AGUqKMCz/Uu1bb8WvtNlEgIMQB5KCvDnnXmB3y498GXe30sbxVcSYADyUFKA2/OQ5sXd+/2g9erkShydzkISYAAyUVSAZ+921i5FOXnSbQICDEAeygrwbPpyOcGTx13viCTAAOShsADPTT8evNzb23t88GaL+xEKMAB5KC/ANyLAAORBgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMMB4NI3VYDYEGGA0mkaB8yHAAKPRrgOtB3MhwACjIcA5EWCA0bALOicCDDAe+psRAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMEKdpRc8EMQQYIE6jwOMlwABhTtZJ1kzjJMAAYQR4zAQYIIwAj5kAA4QR4DETYIA4DsIaMQEGiKO/IybAABBAgAEggAADQAABBoAAAgwAAQQYAAIIMAAEEGAACCDAABBAgAEggAADQAABBoAAAgwA27jhjTQEGAC6u/GdrAQYADq7+Y0kBRgAOurjPs4CDACd3Hjv8+lUepiV1Sn2PcE+CTAAN9NPfgUYADroK78CDADX11t+BRgArqvH/AowAFxPf3ufTyfX37ROp9j3BPskwABspef8CjAA/Fjv+RVgAPih/vMrwADwA0PkV4AB4EoD7H0+nW7vU+x7gn0SYAC6GCq/AgwA3zVcfgUYAL5jyPwKMABsNmh+BRgANhk4vwIMAJcNu/f59Cl6n2LfE+yTAAPwQwnyK8AAsCpJfgUYAFakya8AA8CSVPkVYAAGky5mfUm09/n0uXqfYt8T7FNpiwJAyUoLcMr8CjAAgykrwGnzK8AADKaoACefWQEGgIB/KwgwAGOXeu/z6ZP2PsW+J9gnAQZgTUh+BRiAcQvKrwADMGpxB4oJMACjFXmctgADMFJhe59Pn733KfY9wT4JMAAnYvMrwACMUnR+BRiAMQrPrwADMD4Z5FeAARib+L3Pp7PR+xT7nmCfcnjJAQiUSX4FGIAxySa/AgzAiOSTXwEGYDRyyq8AAzASGe19XhBgAEYgt/wKMAAjkF9+BRiA6uWYXwEGoHZZ5leAAahbpvkVYABqlufe5wUBBqBWGedXgAGoVdb5FWAAKpV3fgUYgCrlnl8BBqBCme99XhBgACpTQn4FGIDKlJFfAQagLoXkV4ABqEkx+RVgAOpRyt7nBQEGoA5F5VeAAahDYfkVYACqUFp+BRiACpSXXwEGoHjF7X1eSBHg6eu+n2NrBb5DAFypzPymCfC3R839N30/zXaKfI8A+K5S85sswHMPcmhwoe8SABuVm99E3wG/3128RJMHb/t+sq6KfZ8AuKzg/CY7CGv68u5Jgx9/6vv5Oin4nQJgVdH5TXkU9FmD//H4776f8vqKfq8AuFDy3ueFpKchfX25s3jBbr2IanDZbxYAp4rPb/rzgE8bPAnaDC797QJgVkV+kwf4/fPm1OSPvp/4Osp/wwCoIb9pA/z+19Md0B+fRxW4hrcMYNzqyG/CAJ9u+57ue54+a5o7fT/zNdTxpgGMVxV7nxfSBPjj6Z7nixOBP+80P/3V91P/WCXvGsBI1ZPflFfCam69WH3shgH+ePC6+znF1bxvACNUU37TBXj9sOf5Y7e3OhD66GQX9snB1CtNv9bM1fPOAYxMXflNFeAN16A82mbaX9pd2bf+mM32zw6m/qXjzNX03gGMSWX5Lex+wJ93Ts9gOp7/7/29vfbHh52mUNm7BzAW1eU30f2APxy8Xtrd/OHlllfhWHyXfO/ufBt49+QkpumfXc9mqu79AxiD2vY+LyT6DnjpgKvtD786bJqf3p5uBz9dPNKezdRpE7i+NxCgelXmt6gAt7VddHce4rMDuI6bbsdy1fgWAlSt0vwOHuB/7839c6eZ/Lx3ZrfZMsDn5Z5vAt9Zf+y6M1flmwhQr1rzO3iATw+bWnNnqwmfx3b+hzvrj1135ip9GwHqVG9+h98Fvb+hv7e2+wp4HtuTI66m/9e9/3K637nrBbXqfSMB6lPt3ueFoQM8PTg4eLXTTH4/OLfVGcCt/ctHXB123Jyu+J0EqEzd+Y04CGt7x5dOOvryyFHQAFWqPb8h5wFvrT0MuvkvF1NaXI+y2wUtK383ASpRf34LuxLWt9UjqBfXmO52V+Hq30+AGowgv4UFuN3mXQ3wrcsXmb7SCN5RgNKNIr9DB/jk29/T2xFe6OlOwNP/3TG/AgyQvTHsfV4oOcDbzNwo3lSAYo0mv4MH+Nd79/9q/3fVfQEG4LIR5be074BvbDxvLEBxRpVfAQYgE+PKb/EBvvoaH5uuQ51w5gC4tvGtoQUYgHBjXEEPfhT0Jv0dBe1uSADlG2N+iw/w7OvRp04zN753GCBz48xvgtOQNnEaEgCnRprf4r8D7mqk7zJArkabXwEGIM5Y9z4vFBjg6dGHg4OD10fb3OFwvG80QHZGnd+gAG/VzhMfny8dzPXgTddfH/NbDZCVkec3WYCnr/bO3Lu7/VHQX3bXjqe+1e12wAIMkIfR5zdVgI93ejkN6WQy905Lfrf9YfK028yN/O0GyIP8Jgrw59X+Ng+22wXdnlU8ebH0wPudrjH3hgPEk99WkgAfttF99aiZ/PeXu503Wlcms5bbNskPO82ctxwgmL3PJ1IEePpskcn9pnm6qOjt7TaA28mst/u449S85wCx5PdMigDPt1MnfyzS+3BzRq9n03WfXQsaoCTyeyFRgNtKzrdW78zO/9OdAAMUTn6XJAzw553F7uL5T9vtg55vO0/WzzqyCxqgGPK7ItF3wG2AT8vbdaP1wv6l2rb7s+90mjnvPUAMe5/XJDkKen+x6XrR4S0D3J7NdPvt0gNf5v29tFF89cx58wEiyO8lqU5DujM7Owy6617j1enMi7v3+0Hr1cmVODqdhSTAABHkd4MkAW5P151Hd57en978n0fbHoQ1927tih7N5EnHmbMAAKQmvxsluhTl4hIa7Te2i2p2vIDzkunL5QRPHnfdlLYIAKQmv5sluhnD+93F17+722y1rpl+PHi5t7f3+ODNFjuyLQQAacnv9yS+HeH7eTk/9f2MHVgMAFKy9/n7Qu4HHMdyAJCO/F4lyXnAHw5eL+0u/vCy81e3vbEkAKQiv1dLeCWsjT8lZlkASER+f0CAAeif/P7QwAH+997cP3eayc97Z3Yv3dU3IcsDQAL2Pl/DwAH+vH7ljIU7fT/ntVkgAAYnv9cy9C7o/Q39vRW2ASzAAEOT32saOsDT9qrNO83k5PrNC0d9P2MHFgqAYcnvdaU/CCuUxQJgSPJ7fQHnAUeyYAAMx97nLpJcCevPWy/6fpYtWTIAhiK/3STZAn52kxsg9cqyATAQ+e3Id8AA3Jz8dpZoC1iAASpm7/MWknwHfBh57Y0VFhCA3snvVtLcjvDdTnPr98jzf89YRAB6Jr9bSrIL+l97/1y5FJZrQQPUQn63leggrEaAASokv9tLEuBf7626L8AAFbD3+SbSfAecDUsKQF/k92YEGIAtyO9NCTAA3cnvjaUL8Nf2VoSvP/X9dN1YXgB6IL89SBXgd3dPj4CePOn7CbuwxADcmL3PvUgU4D+XTkL6JfDWhBYZgBuS356kCfBhu+n74ODg1fOdJvSylBYagBuR394kCfDnnfPt3ul8Wzjw3oQWG4CbkN/+JAnw/vJW737kJrAFB2B78tunRLcjXNronW8O3w77FtiiA7Ate5/7leha0EtXf179KTHLDsB25LdvAgzAD8lv/+yCBuBH5HcADsIC4GryO4gkAT6ev3dnF8B6N//z076f89osQwAd2fs8kDQX4phv9Ta3fj86Onq160IcAAWR38GkCfD02dKlKOO+ARZggG7kdziJrgU9Pb8Y9CTyUtACDNCB/A4p3e0IP/xrb2/vt9d9P103FiWA67L3eVjpApwFyxLA9cjv0AQYgEvkd3gCDMA6+U1g4AB/PdrkU9/PeW2WKIAfkt8khg3wt0fNJq4FDZAte58TEWAALshvMsMH+P7eJb+5GQNAjuQ3oQRbwJPHcd/5rrNgAXyf/KY09FHQH5+37+c/Hkde/mqJRQvge+Q3rQSnIZ00+NaLHBps4QLYzN7n1NKcB/z+1/aNvf+i7+fqzNIFsIn8ppfqQhzT97s5NNjyBXCZ/EZIeCWs6cu77Vv84E3fz9iBJQzgEvkNkfZSlCcNDrwjoWUMYI38Bkl+Legvuy7EAZANe5/DRGwBCzBAHuQ3UMIAfz2pb+iFOSxoABfkN1KqAH99uRNe35kAA1yQ31hJApxJfWcCDHDG3udowwc4n/rOBBjghPzGGzjAX/5nRvWdCTBAS35zkOJ+wLnUdybAADNf/mZi+ADf+m8H6167EAdAFPnNRIot4EucBwwQxN7nbAgwwHjIb0aGDfD0X3ub/GYXNEB68puV5NeCjmXRA8ZLfvMiwACjIL+5EWCAEbD3OT8CDFA9+c2RAANUTn7zJMAAVZPfXAkwQM3kN1sCDFAv+c2YAAPUyt7nrAkwQJ3kN3MCDFAj+c2eAANUSH7zJ8AA1ZHfErgdIUBl7H0ugwADVEV+SzFwgH+9t8l9AQYYhPyWw3fAANWQ35IIMEAt5LcoAgxQB/ktTEiAj/7u+0mvy9IJVMre5+IkCvD01d6Ze3cdBQ3QL/ktUJoAH+84DQlgKPJbpCQB/rza3+aBXdAAvZHfMiUJ8GEb3VePmsl/f7nbNJOnfT/l9VlIgdrIb6lSBHj6rGkezmb7TfN0EePbYRvAAgxUxt7ncqUI8Lf5tu8fi/Q+PKlx3Caw5RSoifyWLFGA28Oujpvmzuz8PzEsqUA95LdsCQP8eWex83n+U9w+aMsqUAv5LV2i74DbAJ+W9zTHMSytQCXkt3hJjoLeX3wHfNFhAQa4EfmtQKrTkO7Mzg6DPo48DNoSC1TA3ucqJAnwfKO3je48vT+9+T+PHIQFcAPyW4lEl6JcXH6yPQOp1e6PDmKhBQonv9VIdDOG97uLr393F/190vdTXp/FFiia/FYk8e0I3+/tPf7U9zN2YMEFSia/NQm5H3Aciy5QLvmtS5LzgD8cvF467vnDy8eOggboyN7n2iS8EtbGnxKz9AJlkt/6CDBA9uS3RgMH+N97c//caSY/753ZbQQYoAv5rdLAAf6802xwp+/nvDbLMFAc+a3U0Lug9zf091bYBrAAA6Wx97laQwd4enBw8Gqnmfx+cO6o72fswGIMFEV+K5b+IKxQFmSgIPJbtYDzgCNZlIFiyG/lXAkLIEvyWzsBBsiQ/NZv2ACffPvb3g54hfOAAa5i7/MYCDBAZuR3HAYO8K/37v/V/u+q+wIM8B3yOxa+AwbIifyOhgAD5EN+R0SAAXJh7/OopAvw9GPwZShbFm0gW/I7MqkC/H739Ajo+2/7fsIuLNxApuR3dNIEePps6SSkh30/YwcWbyBL8jtCaQLc3pRw8vN/Hvy//wwusAUcyJH8jlGSAB/PF61fTm7HMP1znuI/+n7Oa7OIA/mR33FKEuD5BvCdzT+kZiEHcmPv81gluh/w0kbv553mdtjNCS3lQF7kd7wSBXjp6s+rPyVmOQdyIr9jliLA02cCDHCJ/I5bku+AD5vm6fkPx74DBpg59mr0kgT4227z09n1N1Y3h1OztAOZkN/RS3QhjufN5MniT192I89CEmAgD/Y+k+YgrF/v3WuXtX+c/GcSeFdgCzyQAfllluwo6E0i9kRb5IFw8suCAAOkJL+ccj9ggITklzMCDJCM/HJBgAESsfeZZWkD/G6naR68veIvDM2yD0SRX1YlCvD73faQq8PF8uc8YGB85Jd1aQJ8uDjm+fNO3OHPpyz/QAT55bIkAW7LO6/uIsPTZ03zsO/nvDafACCA/LJBqpsx3P67vQp0exsGN2MAxkV+2SjR7Qjb733b7eCnbkcIjIu9z3xHoithnR6C1XZYgIHxkF++K2GA908OvxJgYCzklyukC/DpV8Dtnujbf/f9pNflowAkI79cKdF3wM3Ts6+A2w3hO30/57X5MACpyC9XS3UU9K03/2OxB3r6v5qTDsfwcQDSkF9+JEmAz+5H+PDkT3f6fsrr84EAUrD3mR9LcyWsk2tg3f57EeBfwr4BFmAgBfnlOhJdC3r6fu+3N/P/fvuvD970/YRd+EwAQ5NfrsftCAF6JL9clwAD9Ed+ubZ0Af56MPf6U99P141PBjAg+aWDVAF+d/fkQOhm8qTvJ+zCZwMYjL3PdJIowH82FxwFDVRIfukoTYAP203fBwcHr57vOA8YqJD80lmSALenAZ9u907/PLknUhAfEGAA8ssWkgR45fLPrgUNVEZ+2UaimzEsbfS6GxJQFfllOwnvB7zxp8R8TIB+2fvMtgQYYGvyy/bsggbYkvxyEw7CAtiK/HIzSQJ8PF9Kzy6A9W7+56d9P+e1+bQAPZFfbijNhTjmW73Nrd+Pjo5e7boQB1AB+eXG0gR4+mzpUpRx3wALMNALe5/pQaJrQU/PLwY9ibwUtAADNye/9CLd7Qg//Gtvb++3130/XTc+NMANyS89SRfgLPjYADciv/RGgAGuTX7pjwADXJP80qdhAzxtv/e97DdXwgKKY+8z/Ro2wN8eNZu4FjRQGvmlbwIM8EPyS/8G3gX94eDEq51m8vvBmdd2QQMFkV+GkOggrNB7EC7xGQI6k18GIcAAV5FfBlJYgKcvf733839e7MHuOl2fI6ATe58ZTFkB/vfOyfWkH58lWICBAckvAyoqwIeX7qgkwECvlpMrvwyqpAB/nm//3npxdPRn+9+/tpquDxNwlbN/5M/kl8GVFODDsy3fL7tnBRZgoE8n64j2f+WXoRUU4Omzpnl68cdFiwUY6NHZKsLWLwkUFODlabQFvrPFdH2kgCucriL0lxQKDfDiIpcPBRjolS9/SajUALdHZE3+EGCgT/NVxMVRWDCsgq4FvfQdcOu4aX56K8BAn07ia01BCiXdDenw5HvfpR9/+v8EGOjNxVoqek4Yg5IC3J4H/ODTxc/73SfmYwV8j/ySVqJd0Ku2vR3h4Vpv/xRgoCfKS2qJDsLqybud1d7OfxZg4ObklzNId1YAAB02SURBVPTKCvBs+v63la3n6Z87AgzckPwSobAA35TPGLBOfomRMMDffr13v497At+ETxmwSn6JkjLA/dwT+EZ8zoBl8kscAQZGS36JVHiAr57mpnOQe54BoFhWCcQSYGCUrBCIVnWAL/N5A1ryS7zCAzz7evTpx3/pgk8cIL/kofQAd+QzB8gveRBgYFTkl1y4EhYwIvJLPgoM8PSovcfS66NtbqnkkwdjJr/kpLQAf3y+dErRgzddf91nD8ZLfslLqgB/+Nfehd+2vB/w7Mvu2lm9t/7oOHM+fTBS8ktu0gT4885KNrc9Fut4MZl7pxm/2/4wedpt5nz+YJTkl/wkCfBaf7cN8LdH8+C+WHrg/U7XafkEwhjJLzlKEuDD+bJ///ejc52unbEymbXctkl+2GnmfAZhdOSXPKUI8PRZ09y5+YTbyazvcD5umttdvlD2KYSxkV9ylSLA8+3USceDpTbZdCEP14IGriK/5CtRgPu4BJYAA93ILzlLtAu6jwDPJ3NpQ9ouaOB75Je8pToIq9vZQpvtX6pt52+XfRphNOSXzCUJcE/7oNuzmW6/XXrgy7y/3b5d9nmEkZBfspfsQhyTx0c3nvTh4tIbe78ftF6dXImj01lIAgzjYO8zBUh0EFYvF+KYzd6tXdGjmTzpOHM+klA/+aUIZQV4Nn25nODJ464XlfahhOrJL4VIEuBf7626f5MvhKcfD17u7e09PnizxS0dfCyhcvJLMUq7HeEN+WBC1eSXgggwUAv5pSgCDNRBfimMAAM1kF+KI8BA+eSXAg0b4EvHP/dwFPTNZs5HFOojvxRp4ACvnwF80/OAbzxzPqRQG/mlUAIMlEx+KdbA3wF/PdrkU9/PeW0+qFAV+aVgDsICSiW/FE2AgTLJL4UTYKBE8kvxBBgoj/xSAQEGSiO/VEGAgbLIL5UQYKAk8ks1BBgoh/xSEQEGSiG/VEWAgTLIL5URYKAE8kt1BBjIn/xSIQEGcie/VEmAgbzJL5USYCBjjfxSLQEGsiW/1EyAgUypL3UTYCBHNn6pngAD+ZFfRkCAgdyoL6MgwEBWbPwyFgIMZER+GQ8BBrKhvoyJAAOZkF/GRYCBHNj3zOgIMBBPfhkhAQaiqS+jJMBAKBu/jJUAA4Hkl/ESYCCM+jJmAgzEsPHLyAkwEEF+GT0BBtJTXxBgIDUbv9ASYCAp+YUTAgwkpL5wRoCBVGz8whIBBtKQX1ghwEAK6gtrBBgYnvzCJQIMDMy+Z9hEgIFByS9sJsDAgNQXvkeAgaHY+IUrCDAwDPmFKwkwMAT1hR8QYKB3Nn7hxwQY6Jn8wnUIMNAr9YXrEWCgPwNs/DY2qKmUAAN9GSKVTaPAVEqAgX4MksmzSSow9RFgoAdDbaWeT9OHl+oIMHBjg+0kXpqoTy+1EWDghgb8ilaAqZgAAzcy5BFSyxP26aU2Agxsb+gDlG0BUzEBBjpZOi1o+PODBJiKCTDQRbNq+Kdb/wPUQoCBDpbOy01yaq7zgKmXAAPXd9rBhNemciUsqiXAwPW1H6GzHqYtcJLngpQEGLi+5c1RHye4EQEGrmt1Y9THCW5EgIFrubQr2McJbkSAgR87q68AQ28EGPiB73zv6+MENyLAwJXWvvdtLh4PmiGohAAD33fpFKDzn5wZBDckwMB3bDwDN91VKKFyAgxs8t3K6i/0Q4CBSzQWhifAwCr1hSQEGFiivpCKAANn1BcSEmBgQX0hLQEG1BcCCDCMnvpCBAGGcVNfCCLAMGLqC3EEGHpXStYKmU2olABDzy4ulpz14pb/HELlBBj6dZ61nPtWwD8QoHoCDP1aWsgyXd7UF7IgwNCrtZvnhs3H96gv5EKAoVdZB1h9ISMCDL3KN8DqC3kRYOjT6iKWzwKnvpAdAYZe5bgFrL6QIwGGXmUXYPWFTAkw9CqvAKsv5EuAoV/5nAesvpA1AYZ+5XIlLPWFzAkw9CyHa0GrL+RPgKF3wfmNrj9wLQIMVVFfKIUAQz3UFwoiwFAJ9YWyCDDUQH2hOAIMpQs/5hrYhgBDyeLPeAK2JMBQKvGFogkwFCiDa30ANyTAUBjxhToIMBREe6EeAgyFEF+oiwBDAcQX6iPAkDnxhToJMGRMfKFeAgx5crAzVE6AGYHiUia+MAICTP2aonpW0rwCNyDAVO+iZdlXTXxhRASY2i3nLOe0iS+MjABTu5U3PdMlQHxhhASYyq2959ktAkV9Pw30SICpXM4BFl8YMwGmcrkGWHth7ASYuq2/5VksAuILCDDVy20LWHyBEwJM5XIKsPgCFwT4+r9q3VmkXAJsAQJWCfC1f9MKtFDx5wFbdoANBPiav7d8NcOsh8i62CthiS/wHQJ8zd8r42qGbBJ2LWjtBa4gwFv8mhVqYSK+PxBf4AcE+Fq/tR7g4Qdp/d2npK+mwwWAaxHgbX5r+DWrlXiRGu8bcG0CvM1vDb52DfvWki1JL9CVAG/zW0OvYx3zVRLtBbYiwNv80vABTvhkbE16gRsQ4G1+a+DVrYOu86e9wE0J8Da/JcAjJr1APwR4m98S4HHSXqBHAnyt30p6HnCWd7AdOekFeifAW/xa0mOwBDhWo73AMAT4mr+X8rwgAc6D9AJDEuBr/t7iNxOtjwU4mvQCwxPga//mRn3O3PKTfe8Hhqa9QCICfP1fXVorD1zitHu8OSG9QFICfMPpDRPiiwmJQQraC6QnwH1NuN8SC0Ii0gtEEeC+n6CnEKvC4LQXCCXAgz3ToN8ScxPeFiADAjz4MwpxTrwTQC4EON1TK3EoLz2QFwFOPwtCnJJXG8iUAIdR4gFdfnG9vEBmBDicUvRHd4FyCHA+tGNbo8/uKAcNxRPg/GzajCtixlPzKp0Y+/ihVAKcsc0lHvt61uux5vwFGP0rAYUR4EKMPsYjHvoPrNy5I3A+gI4EuEQjqvEYxngz7l0JpRLg0tUZ48qGMyQBhlIJcFUKr3GxMx5JgKFUAlytcmJcwjxma/Wl8sJBQQR4HHKscVYzUy5bwFAqAR6f78c4SvQrUjQBhlIJ8MjpbukEGEolwFA25wFDoQQYyna+M8FeBSiLAEPh7NWHMgkwFE9+oUQCDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAAFBfjbo41X9P/pry4zJ8AAZEGAASBAQQGefdkVYAAqUVKAZ9NnTfPwRlMQYADyUFSAFwV+epMJCHAm3L4HGL2yAtx+D9xpl/M6K/w8uIEeQGEBnh3fbCe09X0WLrqrwMBolRbg6bMbbQJb3edguboKDIxVaQGefd7b+3+2/21r+xysvAveEmCkigvwzVjbZ2DtTfCeAOMkwKQmwAAzASY9AQaYCTDJrb8H3hNgnASY1GwBA8wEmPQEGGBWfICvvjLWpns3JJw5NhNggJkAE8B5wACVB/gya/scuBIWQPEBnn09+tTlr1vbZ8G1oAGKD3BHVvd58J0AgAATQX+B0RNgAAhQYICnRx8ODg5eH/29xe8KMAB5KC3AH58vfX344E3XXxdgAPJQVoC/7K6d1Xvrj24TEODv8Z0sQFpFBfh4p03Evb0Td9sfJk87TUFgNnOpEoDUSgrwt0fz4L5YeuD9PMidrsMhwN9x3l0BBkikpAAfXsptm+SHXSYhLxstvyxeIoAkCgrw9FnTrO9wPm6a212OhlaXTZrVAHuNABIoKMCbrvvsWtB9WH1VvEYAKQgwAgwQoKAAT581k/WzjuyC7oMAA6RXUIBn+5dq234tfKfLJMRlEwEGSK+kAH/emRf47dIDX+b9vbRRfCVx2USAAdIrKcDteUjz4u79ftB6dXIljk5nIYnLRgIMkF5RAZ6921m7FOXkSbcJiMsmTkMCSK+sAM+mL5cTPHnc9Y5I4rKRC3EAJFdYgOemHw9e7u3tPT54s8X9CNVlM5eiBEitvADfiLxs5mYMAKkJMAvyC5CWAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIUGCAp0cfDg4OXh/9vcXvCjAAeSgtwB+fNxcevOn66wIMQB7KCvCX3WbVrT+6TUCAAchDUQE+3mmje2/vxN32h8nTTlMQYADyUFKAvz2aB/fF0gPv50H+6a8ukxBgAPJQUoAPL+W2TfLDLpMQYADyUFCAp8+aZn2H83HT3O5yNLQAA5CHggI839y9tL9502NXEWAA8iDAABCgoABPnzWT9bOO7IIGoEwFBXi2f6m27dfCd7pMQoAByENJAf68My/w26UHvsz7e2mj+EoCDEAeSgpwex7SvLh7vx+0Xp1ciaPTWUgCDEAmigrw7N3O2qUoJ0+6TUCAAchDWQGeTV8uJ3jyuOsdkQQYgDwUFuC56ceDl3t7e48P3mxxP0IBBiAP5QX4RgQYgDwIMAAEEGAACCDAANSiqHV84QG++lrQzQYJZw6ApMpayQswAJUoayVfdYAvK+q9AaATAU7p69GnLn+9qPcGgIqVHuCOBBiAPAgwAAQQYAAIUGCAp0cfDg4OXh9tcSloAQYgE6UF+OPzpVOKHrzp+usCDEAeygrwl921s3pv/dFtAgIMQB6KCvDx4mbA9/ZO3F3cEvhppykIMAB5KCnA3x7Ng/ti6YH38yB3ug6HAAOQiZICfHgpt22SH3aZhAADkIeCAjx91jTrO5yPm+Z2l6OhBRiAPBQU4E3XfXYtaADKJMAAEKCgAE+fNZP1s47sggagTAUFeLZ/qbbt18J3ukxCgAHIQ0kB/rwzL/DbpQe+zPt7aaP4SgIMQB5KCnB7HtK8uHu/H7RenVyJo9NZSAIMQCaKCvDs3c7apSgnT7pNQIAByENZAZ5NXy4nePK46x2RBBiAPBQW4Lnpx4OXe3t7jw/ebHE/QgEGIA/lBfhGBBiAPAgwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASDA6AIMAHnovXF9T7BP0S82AJzpvXF9TzAfle/Arnt4dY+u8uHVPbrKh1f36LIbXmaz06fcXuqe1T28ukdX+fDqHl3lw6t7dNkNL7PZ6VNuL3XP6h5e3aOrfHh1j67y4dU9uuyGl9ns9Cm3l7pndQ+v7tFVPry6R1f58OoeXXbDy2x2+pTbS92zuodX9+gqH17do6t8eHWPLrvhZTY7fcrtpe5Z3cOre3SVD6/u0VU+vLpHl93wMpudPuX2Uves7uHVPbrKh1f36CofXt2jy254mc1On3J7qXtW9/DqHl3lw6t7dJUPr+7RZTe8zGanT7m91D2re3h1j67y4dU9usqHV/foshteZrPTp9xe6p7VPby6R1f58OoeXeXDq3t02Q0vs9npU24vdc/qHl7do6t8eHWPrvLh1T267IaX2ez0KbeXumd1D6/u0VU+vLpHV/nw6h5ddsPLbHb6lNtL3bO6h1f36CofXt2jq3x4dY8uu+FlNjt9yu2l7lndw6t7dJUPr+7RVT68ukeX3fAym50+5fZS96zu4dU9usqHV/foKh9e3aPLbniZzU6fcnupe1b38OoeXeXDq3t0lQ+v7tFlN7zMZqdPub3UPat7eHWPrvLh1T26yodX9+iyG15ms9On3F7qntU9vLpHV/nw6h5d5cOre3TZDS+z2elTbi91z+oeXt2jq3x4dY+u8uHVPbrshpfZ7ADAOAgwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIUGuAvzzfaZrJg7fR8zGM6ct7TdP8o9bhLXzeaZ5Gz8Mwpu/ad+/ek7+jZ2QQH6v95H179NNfFz/VtoZZGV19K5jVN28hizVMpQF+N/9wtCb/ET0nQzgbXdP8Ej0rg/n2qMng4zGEz2fv3q31NUIFpn+eLZoPo2elb9NnzdI6vLY1zMro6lvBrL55C3msYeoM8HFzLv4l7t3S6Jo70TMzlP0637ul/jbN7fq2gfcvFs3KCjydD+1iHV7bGmZldPWtYFbfvBN5rGGqDHD7b5tbb2ezj7uXX/byLUb3ZnYyvMkf0bMzjONqVm1r2ndv8qTdybdTXaNO/nXR7rf88qy2RXO+CbW0MqltDbMyuvpWMKtv3olM1jBVBvjwbOOifeGrW8kdn/+ztB3enchZGUy7Dsjh49G/w/M1wXGFm8CH5wvkfl2fvPeL/Rbn6/DK1jCro6tuBbP25i3ksoapMcDzxebs323zf5FXt5Lbv1huahxeq/3G5v/O4ePRu6Vlc+mP1dg/H9JxFavuU1/mm4LNg93zdXhda5j10VW2glkf3kI2a5gaAzz/x83ZUlPjSm7J0kirMt/AeHqYw8ejd/M12p3oeRhQpQE+bL82WDqOp641zProltSwgtk4vGzWMDUGePmzv5/DizyYGj4fG8wr9XCWxcejd8c17LH8vpVd0PW8fYeTX/5ePpC2rjXM+uiW1LCC2TS8fNYwlQb4fCWXxYs8mKo2M87NPyzzT32d79xiVItTSP/xJHpeBtB+s7Y4COt5VV9wf22HshrgitYw66NbUsMKZsPwMlrD1Bjg5Re26i2O+equ9N1fm5zsx8zi49G7dmyHFZ8HvPjCrc7RLa3DK1zDbAxwPSuY1eFltIYR4HK1J7fdiZ6J/p2+ZVl8PHo3/+j/8/wcyxpOYFnXnoDUelDP9u+p8QW4ohXMyvByWsMIcLE2nVxegbOvnbL4ePRtcULiYift1z+rucrBssPzf11MatvFProA17SCWR5eVmsYAS7VtLprHZzYP/2kZPHx6NsiwKcL5HGF71/b318+nf7zorK3b2wBrmoFszy8rNYwAlyoL7VcpWbN+ZuXxcejd/tLByeVf/zsuqXL676rZ+vp1MgCXNcKZtObl8UapsYA13WM4mbt5dKrO8xltnyebJ3v3PIFoipZdS+p6/ScNdUeBb2wHuDKVjAXw8tsDVNpgO+c/bm61cCJxa6+6g5zmS1/h1jlFf3b8VUc4OVV2mFto6v2POCFtQDXtoK5GF5ma5gaA1zXdWo2+TN+wRlIZh+P3i2vuqtLVI27Zi9UeyWshdUAV7eCEeCE6rpS6waHNX07syqzj0fv5gvk0sEgVWw7Lalwy/DCUqIqXMOsBLi+FYwAp1TZvUrWzVdzP72NnonBZfENTf/aUytPVtj1Haa0uBvh6ZtW3zHey4mqbw2zPLoKVzCbTnPOYg1TZYDPbmhZy906V9VzeZorZfHx6F/bqHbZ/FrdXr7W/vJpSHei56Zfa6eSVraGWd3BXt0KRoCTOr7YwxD/EvdtdR9KdR+VM1l8PAaw9PZVse9yxcldVk9UdAztwvrFlOpaw6yep1PdCkaA0/r3zunS8x/Rc9K76bMKPx8bZPHxGMLZslnh1RqXl87q/nWxug6vbQ1zMboqVzACnNjihjOTB5V9k9Fa3sio5vOxQRYfj0F8fXl3/r7dfxM9H8P4+Ovik1ff6NbW4ZWtYS5GV+UKRoABgAsCDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAASCAAANAAAEGgAACDAABBBgAAggwAAQQYAAIIMAAEECAoVzT97/uNE3zj9/erj7+/vnd9uEHLy4eOm5WPVz+f27/nWZ+gSUCDMV6t3Pe09tvNz48eXL24HcDPH02/+lp2hkHZgIM5Vpp6k9/nT46/XMltGcbt98N8Oc213fSzz2MngBDob49appbL9q+Hr3cuSjtfrvh+8vR/E9f24fPHj9e3uu87LCZ3G0mf6SZZ+CCAEOhDpe+u203Y5+ePdr8cvbwYmv4pLvfC/A847fffS/OwIAEGAq1v7zdeni6G7ndLF6O6fzxk7/1vQC3j89/6XwPNpCKAEOhVgI83wS+0/73cO2Q5vYQq8X/8b0AL6ay7zAsSE+AoVCHG5La9nY1pcenx2d9J8DzcM+DfewwLEhPgKFQ7fe+v3y69NjazuRvj042lL8T4JOKz7vtMCxITYChVIeLM31//n0pwpcvqnG2Tbw5wGfl3bQ1DQxLgKFY51fcuP/i4mzfO2t/af+kravnAZ/9pbOvji9vOQNDE2Ao1/T97tklrx4vEnzYMcBnR19d/u4YGJoAQ9k+vvz14pJXHbeAL84/urzvGhiYAEP5vjxvzjO73tH9K74DXqmyw7AgLQGGGpx+idvtKOjFfRg2XB4aSEGAoUxrF884+bHbecCfd1YC7DAsSEqAoUxrx1sdnpS305WwDtfuSugwLEhJgKFMq3ubz07oba8FfWfpb111LeizvdPnf9NhWJCSAEOZ2k3WW2/Pfjo824Pc4W5Iq0dszYPuMCxISYChUItvcB+8+fvsdODTvi7uB/z4OvcD3l956HxnNZCGAEOp3m28tsbsz5WHb19cI2stwOtHTB87DAuSEmAo1ufd885O/uPi4fdLDz85e/BygNevmtV+fewwLEhHgKFgH5/fazN7/z//Xnv47vzhfzx4cfHQpQBfPux532FYkJIAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0AAAQaAAAIMAAEEGAACCDAABBBgAAggwAAQQIABIIAAA0CA/x8Gw4TOtJLJCAAAAABJRU5ErkJggg==" width="960" /></p>
<p>SOFA is a validated disease severity scale for the ICU, and generally correlates strongly with mortality. Here, while the mortality rate generally increases as SOFA increases, the smooth fit isn’t necessarily non-decreasing as SOFA values increase. We have added points roughly proportional to the sample size of each SOFA level, and you’ll see towards the high levels of SOFA, very few patients are observed, with the second highest score (13) having a 100% <em>survival</em> rate – although only in two patients!</p>
<p>For binary outcomes, it is often useful to plot the proportion of patients with the outcome (e.g., mortality rate) by the different levels of a covariate of interest. Because sample size plays such an important role in the uncertainty associated with these estimate proportions, it seems appropriate to include an estimate of our uncertainty via a confidence interval.</p>
<p>In the <code>MIMICbook</code> package you installed above, there is a <code>plot_prop_by_level</code> which can plot the proportion of patients with an outcome by one or two factor variables. For instance, if we wished to plot the in hospital mortality rate by the SOFA categories (<code>sofa.cat</code>) we defined above, we can using:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot_prop_by_level(dat2,"sofa.cat","hosp_exp_flg")</code></pre>
<p><img src="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAB4AAAAYACAMAAACgu6WYAAABSlBMVEUAAAAAADoAAGYAOjoAOmYAOpAAZpAAZrYzMzM6AAA6OgA6Ojo6OmY6OpA6Zjo6ZmY6ZpA6ZrY6kLY6kNtNTU1NTW5NTY5Nbm5Nbo5NbqtNjshmAABmOgBmOjpmZjpmZmZmZpBmkJBmkLZmkNtmtttmtv9uTU1ubk1ubm5ubo5ujqtujshuq+SOTU2OTW6Obk2Obm6Oq6uOq8iOq+SOyOSOyP+QOgCQZjqQZmaQkLaQtraQttuQ2/+rbk2rbm6rjm6ryOSr5P+2ZgC2Zjq2Zma2kDq2kGa2kJC2tpC2tra2ttu229u22/+2///Ijk3Ijm7Iq27IyKvI5P/I///bkDrbtmbbtpDb27bb29vb2//b///kq27kyI7kyKvk5Mjk///r6+v/tmb/yI7/25D/27b/29v/5Kv/5Mj/5OT//7b//8j//9v//+T////JvSKVAAAACXBIWXMAAB2HAAAdhwGP5fFlAAAgAElEQVR4nO3d8Z8dBnXm5+sATRzvlqQZpxiykE2D2gVcdtuNN23iQtgmbZpEaYmTLDEUsBO5WNL//2ul0cxoZjS2r9FIR1/O8/wA1oz8QXPec8/LHc3cOTwEAF66w/QfAAA2UsAAMEABA8AABQwAAxQwAAxQwAAwQAEDwAAFDAADFDAADFDAADBAAQPAAAUMAAMUMAAMeKkF/LPd1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsno1HY43/Ud9iRTwmPUDCJFVh6xeTQr4Jgp4zPoBhMiqQ1Ydq7r2Rgp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMC3UcAPvnfn5ORbP7z21o/f/urZP330709ueP/6x8n6AYTIqkNWHQr4Fgr447dPHvvdv7v65rsnZwV8787p+9/8MwV8xfoBhMiqQ1YdCvgWCvjuyVs/fPjRuydv/fTSGx/cPTkr4Afv3vR+BexQhMiqQ1YdCvj5C/jendPnvh+/ffkp7k++e3JewDe+XwE7FCWy6pBVhwJ+/gJ+/6xo3z/5zqW3nXz7x2dv/+DJfz96IvxHCviy9QMIkVWHrDoU8PMX8N2zYv3g/O98HxfwV75/8WvPgD/B+gGEyKpDVh0K+LkL+MG7Z8V6787Vv+T94PrfAV/082+c+dz/WwD8qji81O97DXgBBfzwwd+efhX0ty/eq4AB1lPA1zxXAV/9RqSLAr733dMC/soz3wg8/ex/2PoBhMiqQ1YdPgV9mwV88zPge3ceP/l99DTY3wFftX4AIbLqkFWHAn7xBXz37Kuj7176Ii0F/DOHokRWHbLqUMDPXcA3fhX0pV9/YkGvf5ysH0CIrDpk1aGAn7+Az7//9/L3ASvgI6wfQIisOmTVoYCfv4A/6ft8fQr6M6wfQIisOmTVoYCfv4AfvPv4C5yffa3nDy5eCcsXYd1o/QBCZNUhqw4F/PwF/PCjSz8N6d6di5q9+Dvh90+e+M71f3H6Yx+2fgAhsuqQVYcCvoUCfvjR9x7/vN/T5783FfDD//fxzwP+d34e8DXrBxAiqw5ZdSjg2yjgX9b0xz5s/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwKeLGAA1joonKs8A3551g8gRFYdsurwDFgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FPBtFPCD7905OfnWD6+99eO3v3r+/r999P7/9j8q4KvWDyBEVh2y6lDAt1DAH7998tjv/t3VN989OSvgj568/+TbCviK9QMIkVWHrDoU8C0U8N2Tt3748KN3T9766aU3Prh7clbAD949+coPHz74v0/e/DMFfNn6AYTIqkNWHQr4+Qv43p3T574fv325YH/y3ZPzAv7g7Lnx++fPiBXwE+sHECKrDll1KODnL+DzYn3/5DuX3nby7R8/efujJ8B/9An/5vTHPmz9AEJk1SGrDgX8/AV896xgP7j0DPf9r3z//Ncfv339L4cV8BPrBxAiqw5ZdSjg5y7gB++efer53p0rfwl8XsCP3/zj//7k5FEln/uNM5/7fwuAXxWHl/p9rwEvpIC/9+SroC8+Q62AAdZTwNc8VwFf/VzzWQF/8PgbkH768MHf+iroa9YPIERWHbLq8Cno2yzgG58Bf3D+1Peur4K+av0AQmTVIasOBfzCC/jenU94//rHyfoBhMiqQ1YdCvi5C/jGr4K+UsBnn5m+/inq9Y+T9QMIkVWHrDoU8PMX8Pn3/17+PuBLBXzxDPmDE8+Ar1g/gBBZdciqQwE/fwHf+EpYTwv44u9+714r6PWPk/UDCJFVh6w6FPDzF/CT13q+/lrQTwv43p3HPynJV0E/Y/0AQmTVIasOBfz8BXz+045OnwdffMnVpb8T/uDO6fvffOYVKac/9mHrBxAiqw5ZdSjgWyjghx89fqWNb50+/72pgB+9/1EF/7vrPy94/eNk/QBCZNUhqw4FfBsF/Mua/tiHrR9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVxuGwvoEV8Jj1AwiRVYesKg4HDayAx6wfQIisOmQVcThoYAU8Z/0AQmTVIauGw0EDK+BB6wcQIqsOWTUo4McU8Jj1AwiRVYesGhTwYwp4zPoBhMiqQ1YNCvgxBTxm/QBCZNUhqwj9+zMFPGj9AEJk1SGrCv2rgAetH0CIrDpklaF/FfCc9QMIkVWHrDrW968CnrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBTxm/QBCZNUhqw4FrIDHrB9AiKw6ZNWhgBXwmPUDCJFVh6w6FLACHrN+ACGy6pBVhwJWwGPWDyBEVh2y6lDACnjM+gGEyKpDVh0KWAGPWT+AEFl1yKpDASvgMesHECKrDll1KGAFPGb9AEJk1SGrDgWsgMesH0CIrDpk1aGAFfCY9QMIkVWHrDoUsAIes34AIbLqkFWHAlbAY9YPIERWHbLqUMAKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhTwZAEDsNZB4VzlGfDLs34AIbLqkFWHZ8AKeMz6AYTIqkNWHQpYAY9ZP4AQWXXIqkMBK+Ax6wcQIqsOWXUoYAU8Zv0AQmTVIasOBayAx6wfQIisOmTVoYAV8Jj1AwiRVYesOhSwAh6zfgAhsuqQVYcCVsBj1g8gRFYdsupQwAp4zPoBhMiqQ1YdClgBj1k/gBBZdciqQwEr4DHrBxAiqw5ZdShgBfxSHI43/UflBquWNU5WHc6dAn4pFHDbqmWNk1WHc6eAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsQq0lt8AACAASURBVH5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG6iAx1i+jvXLGiKrDjdQAY+xfB3rlzVEVh1uoAIeY/k61i9riKw63EAFPMbydaxf1hBZdbiBCniM5etYv6whsupwAxXwGMvXsX5ZQ2TV4QYq4DGWr2P9sobIqsMNVMBjLF/H+mUNkVWHG3gbBfzge3dOTr71w2tv/fjtr1761b07b/1UAV9h+TrWL2uIrDrcwFso4I/fPnnsd//u6pvvnlwq4Afvnijgayxfx/plDZFVhxt4CwV89+StHz786FrDPrh7crmA3z9RwNdZvo71yxoiqw438PkL+N6d0+e+H7/95p89feNPvntyuYDv3VHAz7B8HeuXNURWHW7g8xfw+2dF+/7Jdy697eTbP35awA/ePfkP/g74OsvXsX5ZQ2TV4QY+fwHfPfmj0//+4NIz3ve/8v3Lv7578lVfhPUMy9exfllDZNXhBj53AT949+xTz9cr9mkBf3Dy1k8vv/c3znzu/61fLYeX+j1fAK8WN/CaF1HAp387rICfYfmAzdzAaz5tHv/l3I8uv/VSAV/9RqSLAr77+C+HfQr6GT790rF+WV9Rh+NN/1G5gViOLOB//k+XVvnX/ubmAr75GfD7p1//rICfYfk61i/rK0oBt4nluAL+xdcur/LnKuB7d07fr4CfYfk61i9riMdVh6yOK+D3Docv/tu/PPd//dPl9930VdCXfv3+yYXrr5U1/bEPs3wd65c1xOOqQ1ZHFfD9dw6/fnMzP3z6/b+Xvw9YAR/B8nWsX9YQj6sOWR1VwL/42mt/+okFfOMrYT28/ozYp6CfYfk61i9riMdVh6yOLOArf+171YN3T77y7GtBK+DPZPk61i9riMdVh6yOKuD773zKM+CHH136aUhnX3KlgI9g+TrWL2uIx1WHrI4q4IfvHX7nkwv44Uffe9S/3/rpQwX8uVi+jvXLGuJx1SGr4wr40VPgP/yUBv4lTX/swyxfx/plDfG46pDVUQV8/4+/fji89uVvnPn9f7rxdyngz8fydaxf1hCPqw5ZHVXAV1+H4/ApX5GlgI9n+TrWL2uIx1WHrBTwGMvXsX5ZQzyuOmR1VAG/INMf+zDL17F+WUM8rjpkpYDHWL6O9csa4nHVISsFPMbydaxf1hCPqw5ZHV3A//IXbxwOr73xBz/6xN+hgD8fy9exfllDPK46ZHVsAb938SVYn/aSHAr4c7B8HeuXNcTjqkNWRxbw4/79wpe/8fXfvM0Gnv7Yh1m+jvXLGuJx1SGr4wr4w9cPX/rr03/653cOn/a60Ar4eJavY/2yhnhcdcjquAL+weFL569+9ak/G1gBfw6Wr2P9soZ4XHXI6qgCvvLTkD58/UteivI2WL6O9csa4nHVIaujCvjKzwP+1B8OrICPZ/k61i9riMdVh6wU8BjL17F+WUM8rjpkdVQB33/n8M2LX/z84FPQt8Lydaxf1hCPqw5ZHVXAvgjrRbB8HeuXNcTjqkNWxxXwh68fvvhXp//0j7/n25BuieXrWL+sIR5XHbI6roCfvBDWG2+8casvhTX9sQ+zfB3rlzXE46pDVkcW8MO/f/3slShf+8Pb6t/tN83ydaxf1hCPqw5ZHVvAD+//w9cfPQP+8p/c0hdgKWDLF7J+WUM8rjpkdXQBvwDTH/swy9exfllDPK46ZKWAx1i+jvXLGuJx1SGrzyjg+3/8jd//p8f/ednv+z7g22D5OtYva4jHVYesPqOAf/G1w6/9zeP/vMwrYd0Ky9exfllDPK46ZKWAx1i+jvXLGuJx1SGrzyjgF2r6Yx9m+TrWL2uIx1WHrBTwGMvXsX5ZQzyuOmR1VAHf/+NLX3f14df/tS/Cug2W79V0ON70H5UbiKVDVkcVsB9H+CJYvleTAm4TS4esPn8Bf/i6Ar4Vlq9DVh2y6pDVZxXwtS+APuXnAd8Ky9chqw5Zdcjqswr44c+fLeBv3k7/KuDpPwHHklWHrDpk9ZkFfP9//8Y3vv76a1++eB2sf/tXt9S/Cnj6T8CxZNUhqw5ZfWYBP3Z7X3elgJ+yfB2y6pBVh6yOKuAr34akgG+J5euQVYesOmR1VAE//MFv/bUCvm2Wr0NWHbLqkNVRBfyLr93aF14p4AuWr0NWHbLqkNWRBezvgG+f5euQVYesOmR1VAHff+e1P1XAt83ydciqQ1YdsjqqgB++d/jSC/hL4OmPfZjl65BVh6w6ZHVcAf/Lnx8OX7j4VuDb+pLo6Y99mOXrkFWHrDpkdVQBX3s9Sq8FfSssX4esOmTVISsFPMbydciqQ1YdsjqqgF+Q6Y99mOXrkFWHrDpkpYDHWL4OWXXIqkNWCniM5euQVYesOmR1dAH/y1+8cTi89sYf/EgB3xLL1yGrDll1yOrYAn7v4kuwfkcB3w7L1yGrDll1yOrIAn7cv1/48je+/pu32cDTH/swy9chqw5ZdcjquAL+8PXzl8L653cOt/aylNMf+zDL1yGrDll1yOq4Av7B4Uvnr351/53Dryvg22D5OmTVIasOWR1VwFd+GMOHr3/JS1HeBsvXIasOWXXI6qgCvvLjCG/vZxNOf+zDLF+HrDpk1SErBTzG8nXIqkNWHbI6qoDvv3P45sUvfn7wKehbYfk6ZNUhqw5ZHVXAvgjrRbB8HbLqkFWHrI4r4A9fP3zxr07/6R9/z7ch3RLL1yGrDll1yOq4An7yQlhvvPHGrb4U1vTHPszydciqQ1YdsjqygB/+/etnr0T52h/eVv8q4Ok/AceSVYesOmR1bAE/vP8PX3/0DPjLf3JLX4ClgC1fiKw6ZNUhq6ML+AWY/tiHWb4OWXXIqkNWn6uA/4sCvkWWr0NWHbLqkNXRBfwP/+b0b4B/668V8C2xfB2y6pBVh6yOLOD771z8PODfvrW/BZ7+2IdZvg5ZdciqQ1bHFfDj/n3ty//rX/4fX3/9cGuvw6GAp/8EHEtWHbLqkNVxBfze4fDfPHnie//PD5dellIBPwfL1yGrDll1yOqoAn70BPjpq2/8wEtR3g7L1yGrDll1yOqoAv7F1678PGA/DelWWL4OWXXIqkNWRxawH0d4+yxfh6w6ZNUhq6MK+P47V54B+3GEt8LydciqQ1YdsjqqgB++d+nvfd/zd8C3w/J1yKpDVh2yOq6A779z8e2/7x1u6zPQCnj6T8CxZNUhqw5ZHVXA9//48ff/fvkP/vL/fPzf/+obp37/uT8RPf2xD7N8HbLqkFWHrI4q4F987fCs538iPP2xD7N8HbLqkFWHrBTwGMvXIasOWXXI6qgCfkGmP/Zhlq9DVh2y6pCVAh5j+Tpk1SGrDlkp4DGWr0NWHbLqkNXRBfwvf/HG4fDaG3/wIwV8Syxfh6w6ZNUhq2ML+L2LL776nU/6LQr487F8HbLqkFWHrI4s4Mf9+4Uvf+Prv3mbDTz9sQ+zfB2y6pBVh6yOK+APXz986a9P/+mf3zlcel1oBfwcLF+HrDpk1SGr4wr4B4eLH8Bw/x2vBX07LF+HrDpk1SGrowrYT0N6ESxfh6w6ZNUhq6MK2M8DfhEsX4esOmTVISsFPMbydciqQ1YdsjqqgO+/c/jmxS9+fvAp6Fth+Tpk1SGrDlkdVcC+COtFsHwdsuqQVYesjivgD18/fPGvTv/pH3/PtyHdEsvXIasOWXXI6rgCfvJCWG+88catvhTW9Mc+zPJ1yKpDVh2yOrKAH/7962evRPnaH95W/yrg6T8Bx5JVh6w6ZHVsAT+8/w9ff/QM+Mt/cktfgKWALV+IrDpk1SGrowv4BZj+2IdZvg5ZdciqQ1bHFfAPfuuvFfBts3wdsuqQVYesjirgX3zt0vcBK+BbYvk6ZNUhqw5ZHVnAt/XiVwr4KcvXIasOWXXI6qgCvvLDGBTwLbF8HbLqkFWHrI4q4Ifvnf84YAV8eyxfh6w6ZNUhq+MK+F/+/HD4wpe/ceb3vRb0bbB8HbLqkFWHrI4q4F987XCZn4Z0Kyxfh6w6ZNUhKwU8xvJ1yKpDVh2yOqqAX5Dpj32Y5euQVYesOmSlgMdYvg5ZdciqQ1bHFfCPFPDts3wdsuqQVYesPruA7//nxz8I6Qt/eIs/hUEBP2b5OmTVIasOWX1mAX94/nMIv/g3CvhWWb4OWXXIqkNWn1XAp18A/YX/+jcf/eeXbvs58PTHPszydciqQ1YdsvqsAn7vcDh9Gcp/ePRE+LZ/IMP0xz7M8nXIqkNWHbL6jAK+/8557z5q4l9XwLfJ8nXIqkNWHbL67AI++zkMH75+65+Dnv7Yh1m+Dll1yKpDVp9RwL/42vnrXj39JwV8Oyxfh6w6ZNUhKwU8xvJ1yKpDVh2yUsBjLF+HrDpk1SErBTzG8nXIqkNWHbJSwGMsX4esOmTVISsFPMbydciqQ1YdslLAYyxfh6w6ZNUhq88u4GfdVhFPf+zDLF+HrDpk1SErBTzG8nXIqkNWHbL6jAK+/8ffeNbv39IrYk1/7MMsX4esOmTVIavPKOAXavpjH2b5OmTVIasOWSngMZavQ1YdsuqQlQIeY/k6ZNUhqw5ZKeAxlq9DVh2y6pDVZAEvdzDsDFl1yKpDVtd4Bvzy+H9/HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPAYy9chqw5ZdchKAY+xfB2y6pBVh6wU8BjL1yGrDll1yEoBj7F8HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPAYy9chqw5ZdchKAY+xfB2y6pBVh6wU8BjL1yGrDll1yEoBj7F8HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPAYy9chqw5ZdchKAY+xfB2y6pBVh6wU8BjL1yGrDll1yEoBj7F8HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPAYy9chqw5ZdchKAY+xfB2y6pBVh6wU8BjL1yGrDll1yEoBj7F8HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPAYy9chqw5ZdchKAY+xfB2y6pBVh6wU8BjL1yGrDll1yEoBj7F8HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPAYy9chqw5ZdchKAY+xfB2y6pBVh6wU8BjL1yGrDll1yEoBj7F8HbLqkFWHrBTwGMvXIasOWXXISgGPsXwdsuqQVYesFPCUw8H2ZYiqQ1YdslLAQw4HDdwhqQ5ZdchKAc84HDRwiKA6ZPVqOhxv+o/6EingESt3LUxOHbJ6NSngmyjgESt3LUxOHbLqWF0CpxTwCAXcIqcOWXWsLoFTCniEAm6RU4esOlaXwCkFPEP/pgiqQ1Ydu0vgMQU8RP+WSKpDVh3LS+BnCniO/g0RVYesOraXgAIe5FB0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgRupYAffO/Oycm3fnjtrR+//dWzf/rJvz85efOZ96+fvUPRIasOWXWsL4HbKOCP3z557Hf/7uqb756cFfDfnr775M0/U8BXOBQdsuqQVcf6EriNAr578tYPH3707slbP730xgd3T84K+IOTN//jw8fvv97Q22fvUHTIqkNWHetL4BYK+N6d02b9+O3LT3F/8t2TswJ+8O7JHz08ff+T/1bA5xyKDll1yKpjfQncQgG/f/ZM9/2T71x628m3f/zk7R+/ffbM9+6l9yvgnzkUJbLqkFXH+hK4hQK+e/bM9oPzv/N9XMBf+f6VXyvgGzgUHbLqkFXH+hJ4/gJ+8O7Zp57v3bnyl8DXC/jSp6h/48zn/t/61XJ4qd/zxfOQVYesyHpxBfz+018q4FMORYesOmRF1nMV8NUvc75awB/4NqTrfKqsQ1YdsupYXwIv61PQH9x58/rXQK+fvUPRIasOWXWsL4GXVMDv3/D8d/3sHYoOWXXIqmN9Cbygr4K+9uu/vbF/t8/eoeiQVYesOtaXwK18H/B3rvz3MwX84O7JV66/CJYCdihKZNUhq471JfCiXgnrcgHfvfoilQr4jEPRIasOWXWsL4FbKOAH75585dnXgn5awO9/Uv9un71D0SGrDll1rC+B2/hhDB9d+mlI9+5cPA/+4PylKE/OffXavzj9sQ9zKDpk1SGrjvUlcCs/D/ij7z1q12+dPs19toA/OFHAN3MoOmTVIauO9SVwKwX8y5r+2Ic5FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YZh4OwMtaXgAKe4050yKricNDAHetLQAHPcSY6ZBVxOGjgkPUloIDnuBIdsmo4HDRwyfoSUMBzHIkOWTUo4Jb1JaCA5zgSHbJqUMAt60tAAc9xJDpk1aCAW9aXgAKe40h0yCpC/6asLwEFPMeV6JBVhf4tWV8CCniOM9Ehqwz9G7K+BBTwHHeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMehyDgchJUhqo71JaCA5zgUFYeDBu6QVMf6ElDAcxyKiMNBA4cIqmN9CSjgOQ5Fw+GggUvk1LG+BBTwHIeiQQG3yKljfQko4DkORYMCbpFTx/oSUMBzHIoGBdwip471JaCA5zgUEfo3RVAd60tAAc9xKCr0b4mkOtaXgAKe41Bk6N8QUXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6Ehgt4OUOhp0hqw5ZkeUZ8Mvj/6l3yKpDVh3rS8CnoOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLq94PHLwAAG75JREFUkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMBzHIoOWXXIqmN9CSjgOQ5Fh6w6ZNWxvgQU8ByHokNWHbLqWF8CCniOQ9Ehqw5ZdawvAQU8x6HokFWHrDrWl4ACnuNQdMiqQ1Yd60tAAc9xKDpk1SGrjvUloIDnOBQdsuqQVcf6ElDAcxyKDll1yKpjfQko4DkORYesOmTVsb4EFPAch6JDVh2y6lhfAgp4jkPRIasOWXWsLwEFPMeh6JBVh6w61peAAp7jUHTIqkNWHetLQAHPcSg6ZNUhq471JaCA5zgUHbLqkFXH+hJQwHMcig5ZdciqY30JKOA5DkWHrDpk1bG+BBTwHIeiQ1YdsupYXwIKeI5D0SGrDll1rC8BBTzHoeiQVYesOtaXgAKe41B0yKpDVh3rS0ABz3EoOmTVIauO9SWggOc4FB2y6pBVx/oSUMAvx+F4039UbiCWDll1rCqBGyngl0IBt4mlQ1Ydq0rgRgp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEGKuAx6wcQ4qh3yKrDDVTAY9YPIMRR75BVhxuogMesH0CIo94hqw43UAGPWT+AEEe9Q1YdbqACHrN+ACGOeoesOtxABTxm/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEGKuAx6wcQ4qh3yKrDDVTAY9YPIMRR75BVhxuogMesH0CIo94hqw43UAGPWT+AEEe9Q1YdbqACHrN+ACGOeoesOtxABTxm/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEGKuAx6wcQ4qh3yKrDDVTAY9YPIMRR75BVhxuogMesH0CIo94hqw43UAGPWT+AEEe9Q1YdbqACHrN+ACGOeoesOtxABTxm/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEGKuAx6wcQ4qh3yKrDDVTAY9YPIMRR75BVhxuogMesH0CIo94hqw43UAGPWT+AEEe9Q1YdbqACHrN+ACGOeoesOtxABTxm/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEGKuAx6wcQ4qh3yKrDDVTAY9YPIMRR75BVhxuogMesH0CIo94hqw43UAGPWT+AEEe9Q1YdbqACHrN+ACGOeoesOtxABTxm/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEGKuAx6wcQ4qh3yKrDDVTAY9YPIMRR75BVhxuogMesH0CIo94hqw43UAGPWT+AEEe9Q1YdbqACHrN+ACGOeoesOtxABTxm/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAMV8Jj1Awhx1Dtk1eEG3kYBP/jenZOTb/3w2ls/fvurn/r+9bNfP4AQR71DVh1u4C0U8Mdvnzz2u3939c13T776qe9fP/v1Awhx1Dtk1eEG3kIB3z1564cPP3r35K2fXnrjg7sn5wV84/sVsOULcdQ7ZNXhBj5/Ad+7c/rc9uO33/yzp2/8yXdPzgv4xvcrYMtX4qh3yKrDDXz+An7/rGjfP/nOpbedfPvHF29/9v0K+GeWr8RR75BVhxv4/AV89+SPTv/7g/NPOT8u2698/+LXN71fAf/M8pU46h2y6nADn7uAH7x79qnle3eu/iXvWeHe8P7fOPO5/7dgxuGlfn8ez0NWZClgeJaj3iErsp6rgK9+o9GzBXz9G5Gmn/0PWz+AEJ/W7JBVhxt4mwV87DNgBXxq/QBCHPUOWXW4gQp4zPoBhDjqHbLqcAOfu4A/8aucfRX0Z1g/gBBHvUNWHW7g8xfw+ff3Xv8+3w+uff+v7wO+Zv0AQhz1Dll1uIHPX8Cf9EpXH3glrE+3fgAhjnqHrDrcwOcv4Afvnnzlptd6Pi/gT3r/+tmvH0CIo94hqw438PkL+OFHl37a0b07F89zL/7O9yM/DelG6wcQ4qh3yKrDDbyFAn740fce9eu3Tp/f3lTAl9+vgJ9aP4AQR71DVh1u4G0U8C9r+mMftn4AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij3iGrDjdQAY9ZP4AQR71DVh1uoAIes34AIY56h6w63EAFPGb9AEIc9Q5ZdbiBCnjM+gGEOOodsupwAxXwmPUDCHHUO2TV4QYq4DHrBxDiqHfIqsMNVMBj1g8gxFHvkFWHG6iAx6wfQIij/mo6HG/6j8oN3EAFPGb9AELc71eTAm5zAxXwmPUDCHG/OzyuOmSlgMesH0CIAu7wuOqQlQIes34AIQq4w+OqQ1YKeMz6AYQo4A6Pqw5ZKeAx6wcQooA7PK46ZKWAx6wfQIgC7vC46pCVAh6zfgAhCrjD46pDVgp4zPoBhCjgDo+rDlkp4DHrBxCigDs8rjpkpYDHrB9AiALu8LjqkJUCHrN+ACEKuMPjqkNWCnjM+gGEKOAOj6sOWSngMesHEKKAOzyuOmSlgMesH0CIAu7wuOqQlQIes34AIQq4w+OqQ1YKeMz6AYQo4A6Pqw5ZKeAx6wcQooA7PK46ZKWAx6wfQIgC7vC46pCVAh6zfgAhCrjD46pDVgp4zPoBhCjgDo+rDlkp4DHrBxCigDs8rjpkNVnAEHHwwABeNM+AX571AwjxDLjD46pDVgp4zPoBhCjgDo+rDlkp4DHrBxCigDs8rjpkpYDHrB9AiALu8LjqkJUCHrN+ACEKuMPjqkNWCnjM+gGEKOAOj6sOWSngMesHEKKAOzyuOmSlgMesH0CIAu7wuOqQlQIes34AIQq4w+OqQ1YKeMz6AYQo4A6Pqw5ZKeAx6wcQooA7PK46ZKWAx6wfQIgC7vC46pCVAh6zfgAhCrjD46pDVgp4zPoBhCjgDo+rDlkp4DHrBxCigDs8rjpkpYDHrB9AiALu8LjqkJUCHrN+ACEKuMPjqkNWCnjM+gGEKOAOj6sOWSngMesHEKKAOzyuOmSlgMesH0CIAu7wuOqQlQIes34AIQq4w+OqQ1YKeMz6AYQo4A6Pqw5ZKeAx6wcQooA7PK46ZKWAx6wfQIgC7vC46pCVAh6zfgAhCrjD46pDVgp4zPoBhCjgDo+rDlkp4DHrBxCigDs8rjpkpYDHrB9AiALu8LjqkJUCHrN+ACEKuMPjqkNWCnjM+gGEKOAOj6sOWSngMesHEKKAOzyuOmSlgMesH0CIAu7wuOqQlQIes34AIQq4w+OqQ1YKeMz6AYQo4A6Pqw5ZKeAx6wcQooA7PK46ZKWAx6wfQIgC7vC46pCVAh6zfgAhCrjD46pDVgp4zPoBhCjgDo+rDlkp4DHrBxCigDs8rjpkpYDHrB9AiALu8LjqkJUCHrN+ACEKuMPjqkNWCnjM+gGEKOAOj6sOWSngMesHEKKAOzyuOmSlgMesH8Ar6nC86T8qN/C46pCVAh6zfgCvKAXc5nHVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVISsFPGb9AEJk1SGrDlkp4DHrBxAiqw5ZdchKAY9ZP4AQWXXIqkNWCnjM+gGEyKpDVh2yUsBj1g8gRFYdsuqQlQIes34AIbLqkFWHrBTwmPUDCJFVh6w6ZKWAx6wfQIisOmTVIavJAl7uN35j+k/AsWTVIasOWV2jgF8ey9chqw5ZdcjqGgX88li+Dll1yKpDVtco4JfH8nXIqkNWHbK6RgG/PJavQ1YdsuqQ1TUK+OWxfB2y6pBVh6yuUcAvj+XrkFWHrDpkdY0CfnksX4esOmTVIatrFDAADFDAADBAAQPAAAUMAAMUMAAMUMAAMEABA8AABQwAAxTwS/Lge3dOTr71w+k/BlfdPTk5+eqTf/z47a/e8Bt+8t2Tkzf/p58+/sd7d/7oJf7JuO48q0uRXPHx2yenfvfvZDXqLKjLeVxy9XF2785bP73yMFxGAb8cZ8t4bReZdumRf/emE/D+kxvyldPc7opv0llWVyK57N6dSwdfVnPOgrqSx+X3XnqcPXj3RAHz4t09eeuHDz96sm28Ou6enD1TenD3phNw786b//Hhw4++++RdNz9H5iV5ktXVSC774PKbZDXn7EH1wU2deu1x9uj/TT05iTf+5gUU8Etx787p/w38+O03/2z6j8Jl5wX8+NOaN5yAuyffefxfZ/E9fF98g55kdS2SK+//zqVfyWrM2YPqah5PXHucPX6SrIB54d4/W6/3b9hJBp3dikf/R/zbP/7kE/Dx20+u/b07O6/Eq+HisxWPnUfy1IN3r1SurMY8CepaHqeuPc4evHvyH+4oYF6489uxdc1eWecF/JXvf1o2986uxE03hZflSgGfR/LUx2+/9f88en71786+0FFWY54EdS2PU9ceZ3dPvnpPAfPCXRyDZ88Goy4d9U8+AT++c/67fAZj0OUCfhrJhfOv+ZHVtPO/rL+ax7mnj7MPTt76qQLmxVPAr6ojCvjuycmb3/+M38NL8DSry5Fc+ODk5Ns/ffj/fe/k7LEmqylPgrqex7mLXE6/IkYB8+JdKmDfHfFK+ewCfvC//Q93Tt78n5/8Qn6Dnn7F+uVILpx/ncX51/7IasqToK7nce7icXb6DgXMi+cZ8KvqqE9BP/zJ+Sc8n/3SH16aK38H/JNnPwd95oOzr6uV1ZQrQV3k8fTX51+Q+vjtCpgXTwG/qo4r4EtH3Vf2jPn0u37h/JmvrKZcDer6ZyLOHmf37pzGo4B5CXwV9CvqyAI+PxOO+qDrd/0TC1hWsz49qLPH2dkLmp2/VNbWy6iAX4rzr8j0lZmvmE8v4Afvnr376VH3ac0xp1k9E8m5i3ec5yirKVeDuv64UsCXKeCXwithvaI+4xnw3YsvJHny376wZ9D5K2FdjeTSu0/fcKmhZTXjSlAXeZy7+jjzKWhegkdb+BWvBf0K+owCvnfn8bdSPPhb39ryCrj49tIrkVx48o6Pvnv+GJPVlMtBPc3jnAK+TAG/HB/5aUivpBsK+OyrQ87feOpNL+4w7+I1/p9GciWrs09q/u4Pz38pqxlPX9/1Io9LQSngyxTwS/LR9x7t4rc8/33FfFYBP/zo3z+69d/y8oavgPOsLkVyQ1bfPnuMyWrMlaC+ffaztBXwjRQwm9294btJP/4fP+lye4H/SbKK+HxBPaGAYZ+bbsX7n/gXBXc9qRokq4jPF9QTChj2ueFWPPjfPuly+yHvo2QV8bmCOqOAYZ+7V34++Gf9Zl9DN0lWEZ8rqF/y3/hVoYDZ7PM88u/d+YRXH+alkFWEAj6eAgaAAQoYAAYoYAAYoIABYIACBoABChgABihgABiggAFggAIGgAEKGH5V3f+L1w+HX/vT6T8GcDMFDL+qfnB45Nf+5pf91//+X//TLf5hgOsUMPyK+vD1w3/1HBX6g8OXFDC8SAoYfkX9/HD45nP86woYXjAFDL+ifn547Xn+/lcBwwumgOFXlAKGV5sChqx//k+vHw6Hf/UnV3792m/99eN/fu9w6rSD//H0973xB9cL9fT3P/3XL/+2s3/9d17KxwE7KWCo+vvDmS/+zemvz0rz8Np/9/ByAd//z9d+37nz3//bp7+6+tsUMLxwChiiPnz98KVHT3bv//nh8OuPf/2oM7/4V4+ex/7e2RdfnX8K+tHbf/vRk9p//r2z33fu0dsf/fuPe/d3bvptPgUNL5gChqj3zr/H9wen//CLr5015v13nrzjrIAfvf1J7178hodXfnn225/5bQoYXjAFDFHvXf0iq6e/fPTU+PFz2rMCvvharPNiPnPxXUpPfsMzv00BwwumgCHq54+/4OqvLn75g4t6PXsy+8xXQf/gSgG/90kvkvUDBQwvhQKGqidfNfWFP/jR4188euJ6Xphn/3i5gP/lH//yj3/z6utS3lSwl3+bAoYXTAFD1j/8m7MvW/6T09a9+BKrH1wt4L//zfOvb/7UAr722xQwvGAKGMr+8Y9Pa/Obn/wM+NGvHn937zf+lx/94NMK+JnfpoDhBVPAEPf4+5AedeUn/R3w2bcXPbz+d8AXv3ry3PmZ36aA4QVTwNB06XPOp5X53sXPXvj54dJXQT99Zvyol2/8KugPX7/yBPr8tylgeMEUMES9d+kbhx515Sd8H/DTZn3v6t8BX/z+0/p+9rcpYHjBFDBEPWrQ1/7w8WtXvfPkueylV8I6fWmrq5+CfvzmizecvVLW5VfCuuG3vfan93809tHBrz4FDFUfvn7+ZctPXrL5ymtBP30lrN87e/Nv/flFUX/zyu8/fS3oZ37bzx//4tdv+t8FboUChqz7f/HG428E/u3z56lPfhrS+S/Pvw3p/l/85pOX7Dj74qynf1l85achXf9tD//+0Tt/3Weh4YVRwAAwQAEDwAAFDAADFDAADFDAADBAAQPAAAUMAAMUMAAMUMAAMEABA8AABQwAAxQwAAxQwAAwQAEDwAAFDAADFDAADFDAADBAAQPAAAUMAAMUMAAM+P8B6ljqDh+yn4YAAAAASUVORK5CYII=" width="960" /></p>
<p>Often it’s useful to consider more than one covariate at a time to assess confounding and effect modification. Here, if we wished to examine <code>sofa.cat</code> and <code>gender_num</code> at the same time, we add <code>factor.var2="gender_num"</code> to our previous use of <code>plot_prop_by_level</code>.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot_prop_by_level(dat2,"sofa.cat","hosp_exp_flg",factor.var2="gender_num")</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@a06fa37fd60e4c37b97191fdeaac6e40" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Questions</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@5292d107698e48c590036c2c53ce940b">
<div class="xblock xblock-public_view xblock-public_view-problem xmodule_display xmodule_ProblemBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@5292d107698e48c590036c2c53ce940b" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="problem" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="True" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "Problem"}
</script>
<div id="problem_5292d107698e48c590036c2c53ce940b" class="problems-wrapper" role="group"
aria-labelledby="5292d107698e48c590036c2c53ce940b-problem-title"
data-problem-id="block-v1:MITx+HST.953x+3T2020+type@problem+block@5292d107698e48c590036c2c53ce940b" data-url="/courses/course-v1:MITx+HST.953x+3T2020/xblock/block-v1:MITx+HST.953x+3T2020+type@problem+block@5292d107698e48c590036c2c53ce940b/handler/xmodule_handler"
data-problem-score="0"
data-problem-total-possible="0"
data-attempts-used="0"
data-content="
<h3 class="hd hd-3 problem-header" id="5292d107698e48c590036c2c53ce940b-problem-title" aria-describedby="block-v1:MITx+HST.953x+3T2020+type@problem+block@5292d107698e48c590036c2c53ce940b-problem-progress" tabindex="-1">
Multiple Choice
</h3>
<div class="problem-progress" id="block-v1:MITx+HST.953x+3T2020+type@problem+block@5292d107698e48c590036c2c53ce940b-problem-progress"></div>
<div class="problem">
<div>
<p>a) Make a factor plot of the categories of SOFA we created (sofa.cat) and hospital mortality (hosp_exp_flg). Does the trend align with your expectations based on the non-graphical EDA performer earlier?</p>
<p>b) Use plot_prop_by_level using sofa.cat as the covariate of interest and 28 day mortality (day_28_flg) as the outcome</p>
<p>c) Include the main covariate of interest for this study <strong>aline_flg</strong> as the second factor variable and extend part b)</p>
<p>d) Repeat part c), but swap the IAC and SOFA arguments. Consider how the different depictions of the underlying data could better support different objectives </p>
<p>e) Create a new variable, <strong>sofa.cat2</strong>, with cut points at 3, 6, 9, 12. Repeat parts b) and c).</p>
<p>f) Make a plot of the 28 day mortality outcome, <strong>aline_flg</strong> and <strong>chf_flg</strong>. Ignoring the statistical significance (i.e., do not perform any formal testing), consider why this plot may suggest the complexity of any potential effect of an IAC on mortality.ke this.</p>
</div>
<div class="action">
<input type="hidden" name="problem_id" value="Multiple Choice" />
<div class="submit-attempt-container">
<button type="button" class="submit btn-brand" data-submitting="Submitting" data-value="Submit" data-should-enable-submit-button="True" aria-describedby="submission_feedback_5292d107698e48c590036c2c53ce940b" >
<span class="submit-label">Submit</span>
</button>
<div class="submission-feedback" id="submission_feedback_5292d107698e48c590036c2c53ce940b">
<span class="sr">Some problems have options such as save, reset, hints, or show answer. These options follow the Submit button.</span>
</div>
</div>
<div class="problem-action-buttons-wrapper">
</div>
</div>
<div class="notification warning notification-gentle-alert
is-hidden"
tabindex="-1">
<span class="icon fa fa-exclamation-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="5292d107698e48c590036c2c53ce940b-problem-title">
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification warning notification-save
is-hidden"
tabindex="-1">
<span class="icon fa fa-save" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="5292d107698e48c590036c2c53ce940b-problem-title">None
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
<div class="notification general notification-show-answer
is-hidden"
tabindex="-1">
<span class="icon fa fa-info-circle" aria-hidden="true"></span>
<span class="notification-message" aria-describedby="5292d107698e48c590036c2c53ce940b-problem-title">Answers are displayed within the problem
</span>
<div class="notification-btn-wrapper">
<button type="button" class="btn btn-default btn-small notification-btn review-btn sr">Review</button>
</div>
</div>
</div>
"
data-graded="False">
<p class="loading-spinner">
<i class="fa fa-spinner fa-pulse fa-2x fa-fw"></i>
<span class="sr">Loading…</span>
</p>
</div>
</div>
</div>
</div>
</div>
<div class="xblock xblock-public_view xblock-public_view-vertical" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@vertical+block@fe572be1b1664a97a0c2f729088b84ae" data-init="VerticalStudentView" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="vertical" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<h2 class="hd hd-2 unit-title">Odds Ratios</h2>
<div class="vert-mod">
<div class="vert vert-0" data-id="block-v1:MITx+HST.953x+3T2020+type@html+block@004bb3cfd17445248b19f03005c2d75a">
<div class="xblock xblock-public_view xblock-public_view-html xmodule_display xmodule_HtmlBlock" data-usage-id="block-v1:MITx+HST.953x+3T2020+type@html+block@004bb3cfd17445248b19f03005c2d75a" data-init="XBlockToXModuleShim" data-graded="False" data-request-token="f220712e43c311ef8f150e08775edbcd" data-block-type="html" data-runtime-version="1" data-course-id="course-v1:MITx+HST.953x+3T2020" data-has-score="False" data-runtime-class="LmsRuntime">
<script type="json/xblock-args" class="xblock-json-init-args">
{"xmodule-type": "HTMLModule"}
</script>
<p><em>Note: For those with a programming background, <code>R</code> indexes vectors starting from 1.</em></p>
<p>As discussed earlier today, odds ratios are very commonly used to communicate relative effect sizes for binary outcomes, particularly in observational data. Calculation is straightforward, but often misunderstood. We start with a 2 x 2 table. Below is the 2 x 2 table for in hospital mortality and having an arterial line. I’ve assigned it to a new variable called <code>egtab</code>.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>egtab <- table(dat2$aline_flg,dat2$hosp_exp_flg,dnn=c("Aline","Hosp. Mort"))
egtab</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code> Hosp. Mort
Aline 0 1
0 702 90
1 830 154</code></pre>
<p>It’s hard to interpret the raw counts, so we’ll use <code>prop.table</code> to compute the proportions who died and lived by row (margin 1, aline).</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>pegtab <- prop.table(egtab,1)
pegtab</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code> Hosp. Mort
Aline 0 1
0 0.8863636 0.1136364
1 0.8434959 0.1565041</code></pre>
<p>Odds are <span class="math inline">\(\frac{p}{1-p}\)</span> where <span class="math inline">\(p\)</span> is the proportion with the outcome (death) in a group of patients, which is in the second column. We can index the above table by column (<code>tab[,idx]</code> will retrieve column <code>idx</code> from the table [or matrix] <code>tab</code>) to compute the odds in each group.</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>Oddsegtab <-pegtab[,2]/pegtab[,1]
Oddsegtab</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>0 1
0.1282051 0.1855422</code></pre>
<p>Now we have the odds of the outcome in those who got an aline <code>1</code> and those who didn’t <code>0</code>. We need to pick a reference group. We’ll calculate it both ways, but let’s assume we want those without an aline to be the reference:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>Oddsegtab[2]/Oddsegtab[1]</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>1
1.447229</code></pre>
<p>If we wanted those with an aline to be the reference group:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>Oddsegtab[1]/Oddsegtab[2]</code></pre>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #FFFFFF; border: 1px solid #cccccc;"><code>0
0.6909757</code></pre>
<p>If we wanted to plot this information, and include a confidence interval, we can use the <code>plot_OR_by_level</code> from the <code>MIMICbook</code> package:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot_OR_by_level(dat2,
"aline_flg",
"hosp_exp_flg")</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
<p>This by default includes an odds ratio of 1 indicating the reference group. To remove this point use the <code>include.ref.group.effect</code> argument:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot_OR_by_level(dat2,
"aline_flg",
"hosp_exp_flg",
include.ref.group.effect = FALSE)</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
<p>You can also look at more than one covariate at a time. For instance, looking at <code>aline_flg</code> and the <code>gender_num</code> variable:</p>
<pre class="example" style="box-sizing: border-box; overflow: auto; font-family: 'courier new', monospace; border-radius: 4px; padding: 9.5px; margin-top: 0px; font-size: 14.4px; line-height: 1.42857; word-break: break-word; color: #333333; background-color: #f5f5f5; border: 1px solid #cccccc;"><code>plot_OR_by_level(dat2,
"gender_num",
"hosp_exp_flg",
factor.var2="aline_flg",
include.ref.group.effect = TRUE)</code></pre>
<p><img src="data:image/png;base64,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" width="960" /></p>
<p>Here we have computed the odds ratio for aline (vs no aline) separately for men and women.</p>
</div>
</div>
</div>
</div>
© All Rights Reserved