update chapter imputation

smartdata-analysis-and-statistics · Mar 11, 2024 · 5cbd311 · 5cbd311
1 parent 72bbb25
commit 5cbd311
Show file tree

Hide file tree

Showing 38 changed files with 8,047 additions and 197 deletions.
diff --git a/_freeze/chapter_09/execute-results/html.json b/_freeze/chapter_09/execute-results/html.json
diff --git a/_freeze/chapter_09/figure-html/homimp convergency_ggmice-1.png b/_freeze/chapter_09/figure-html/homimp convergency_ggmice-1.png
diff --git a/chapter_09.qmd b/chapter_09.qmd
@@ -94,6 +94,7 @@ fplot <- plot_dis(data = data_hom)  #Age density
 fplot
 ```
 
+
 We can calculate the treatment effect on the complete observed dataset. To do this, we start by balancing the dataset using Propensity Score matching. In this case, the propensity score model uses confounder variables only: `age`, `gender`, `prevDMTefficacy`, `logPremedicalcost`, and `prerelapseNum`.
 
 ```{r full1}
@@ -267,7 +268,7 @@ method["numSymptoms"] <- "pmm"
 method["prevDMTefficacy"] <- "pmm"
 
 # Generate 10 imputed datasets
-imp <- mice(impdata, form = form_y, method = method, m = 10, maxit = 10,  
+imp <- mice(impdata, form = form_y, method = method, m = 100, maxit = 20,  
             printFlag = FALSE)
 ```
 
@@ -277,6 +278,16 @@ Before proceeding with the estimation procedure, we inspect the convergence of a
 plot(imp)
 ```
 
+We can also inspect these plots using **ggmice**:
+
+```{r homimp convergency_ggmice}
+#| message: false
+#| warning: false
+require(ggmice)
+plot_trace(imp, c("age", "numSymptoms", 
+                  "logPremedicalcost", "prerelapseNum", "prevDMTefficacy"))
+```
+
 As there don't seem to be any problems in the trace plot (i.e., no marked tendency & well-dispersed plots), we can now proceed with the PS Matching in each of the imputed datasets using the **MatchedThem** package functions. Here, we adopt full matching without replacement and use a logistic model. The function allows the pooling of PS estimates, which can be done through the within or across approach.
 
 
@@ -404,6 +415,25 @@ ggpubr::ggarrange(Relapserate_plot0(data = data_hom),
                   common.legend = TRUE, legend = "bottom")
 ```
 
+A summary of the complete data is given below
+
+```{r}
+#| echo: false
+#| warning: false
+df.tbl <- m_data_het %>% mutate(premedicalcost = exp(logPremedicalcost))
+
+label(df.tbl$age) <- "Age"
+units(df.tbl$age) <- "years"
+label(df.tbl$gender) <- "Gender"
+label(df.tbl$prerelapseNum) <- "Previous number of relapses"
+label(df.tbl$prevDMTefficacy) <- "Efficacy of previous disease modifying therapy"
+label(df.tbl$premedicalcost) <- "Previous medical cost (USD)"
+units(df.tbl$premedicalcost) <- "\\$"
+label(df.tbl$numSymptoms) <- "Previous number of symptoms"
+
+table1(~ age + gender + prerelapseNum + prevDMTefficacy + premedicalcost + numSymptoms | treatment, data = df.tbl, caption = "Baseline characteristics of the case study data")
+``` 
+
 We use also the G computation to calculate the treatment effect, but this time in the main model, we take into account the interaction of the treatment and the covariates. In addition, we will estimate the conditional average treatment effect given the Iscore groups.
 $$\tau_{CATE}(x) =E(y^1-y^0|X=x)$$
 

diff --git a/chapter_09_files/figure-html/homimp convergency_ggmice-1.png b/chapter_09_files/figure-html/homimp convergency_ggmice-1.png
diff --git a/docs/authors.html b/docs/authors.html
@@ -2,7 +2,7 @@
 <html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"><head>
 
 <meta charset="utf-8">
-<meta name="generator" content="quarto-1.3.433">
+<meta name="generator" content="quarto-1.3.450">
 
 <meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
 
@@ -253,33 +253,65 @@ <h1 class="title"><span class="chapter-number">11</span>&nbsp; <span class="chap
 <td>Utrecht University, Utrecht, The Netherlands</td>
 </tr>
 <tr class="even">
+<td>Christine Fletcher</td>
+<td>GlaxoSmithKline, Stevenage, United Kingdom</td>
+</tr>
+<tr class="odd">
 <td><a href="https://orcid.org/0000-0002-5263-3964">Elvira D’Andrea</a></td>
 <td>AbbVie Inc., Boston, MA, United States</td>
 </tr>
-<tr class="odd">
+<tr class="even">
 <td><a href="https://orcid.org/0000-0001-5705-8615">Fabio Pellegrini</a></td>
 <td>Biogen Spain, Madrid, Spain</td>
 </tr>
-<tr class="even">
+<tr class="odd">
 <td><a href="https://orcid.org/0000-0001-9310-6274">Gabrielle Simoneau</a></td>
 <td>Biogen Canada, Toronto, Canada</td>
 </tr>
+<tr class="even">
+<td><a href="https://orcid.org/0000-0001-9767-1924">Gerko Vink</a></td>
+<td>Utrecht University, Utrecht, The Netherlands</td>
+</tr>
 <tr class="odd">
 <td><a href="https://orcid.org/0000-0001-5536-8038">Grammati Sarri</a></td>
 <td>Cytel Inc., London, United Kingdom</td>
 </tr>
 <tr class="even">
+<td><a href="https://orcid.org/0000-0003-3276-2141">Hanne Oberman</a></td>
+<td>Utrecht University, Utrecht, The Netherlands</td>
+</tr>
+<tr class="odd">
 <td><a href="https://orcid.org/0000-0002-6154-8091">Jamie Elvidge</a></td>
 <td>National Institute for Health and Care Excellence (NICE), Manchester, United Kingdom</td>
 </tr>
+<tr class="even">
+<td><a href="https://orcid.org/0000-0001-5531-841X">Janie Coulombe</a></td>
+<td>Université de Montréal, Montréal, Canada</td>
+</tr>
 <tr class="odd">
 <td>Jeremy Dietz</td>
 <td>National Institute for Health and Care Excellence (NICE), London, United Kingdom</td>
 </tr>
 <tr class="even">
+<td><a href="https://orcid.org/0000-0003-2420-5716">Jessica L. Rohmann</a></td>
+<td>Institute of Public Health, Charité, Universitätsmedizin Berlin, Berlin, Germany</td>
+</tr>
+<tr class="odd">
+<td><a href="https://orcid.org/0000-0002-2384-5415">Johanna Muñoz</a></td>
+<td>Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands</td>
+</tr>
+<tr class="even">
+<td><a href="https://orcid.org/0000-0001-7061-3767">Katrine Strandberg-Larsen</a></td>
+<td>Department of Public Health, University of Copenhagen, Copenhagen, Denmark</td>
+</tr>
+<tr class="odd">
 <td><a href="https://orcid.org/0000-0001-9718-021X">Konstantina Chalkou</a></td>
 <td>Institute of Social and Preventive Medicine (ISPM), University of Bern, Bern, Switzerland</td>
 </tr>
+<tr class="even">
+<td><a href="https://orcid.org/0000-0002-6479-4757">Marian Mitroiu</a></td>
+<td>Biogen International GmbH, Baar, Zug, Switzerland</td>
+</tr>
 <tr class="odd">
 <td><a href="https://orcid.org/0000-0002-0346-2871">Mohammad Ehsanul Karim</a></td>
 <td>School of Population and Public Health, University of British Columbia, Vancouver, Canada</td>
@@ -288,6 +320,10 @@ <h1 class="title"><span class="chapter-number">11</span>&nbsp; <span class="chap
 <td><a href="https://orcid.org/0000-0002-6887-5706">Tammy Jiang</a></td>
 <td>Biogen, Cambridge, MA, United States</td>
 </tr>
+<tr class="odd">
+<td><a href="https://orcid.org/0000-0001-6485-9947">Tat Thang Vo</a></td>
+<td>The Wharton School, The University of Pennsylvania, Philadelphia, Pennsylvania, United States</td>
+</tr>
 </tbody>
 </table>