Five steps for missing data with Finalfit

As a journal editor, I often receive studies in which the investigators fail to describe, analyse, or even acknowledge missing data. This is frustrating, as it is often of the utmost importance. Conclusions may (and do) change when missing data is accounted for.  A few seem to not even appreciate that in conventional regression, only rows with complete data are included.

These are the five steps to ensuring missing data are correctly identified and appropriately dealt with:

  1. Ensure your data are coded correctly.
  2. Identify missing values within each variable.
  3. Look for patterns of missingness.
  4. Check for associations between missing and observed data.
  5. Decide how to handle missing data.

Finalfit includes a number of functions to help with this.

Some confusing terminology

But first there are some terms which easy to mix up. These are important as they describe the mechanism of missingness and this determines how you can handle the missing data.

Missing completely at random (MCAR)

As it says, values are randomly missing from your dataset. Missing data values do not relate to any other data in the dataset and there is no pattern to the actual values of the missing data themselves.

For instance, when smoking status is not recorded in a random subset of patients.

This is easy to handle, but unfortunately, data are almost never missing completely at random.

Missing at random (MAR)

This is confusing and would be better stated as missing conditionally at random. Here, missing data do have a relationship with other variables in the dataset. However, the actual values that are missing are random.

For example, smoking status is not documented in female patients because the doctor was too shy to ask. Yes ok, not that realistic!

Missing not at random (MNAR)

The pattern of missingness is related to other variables in the dataset, but in addition, the values of the missing data are not random.

For example, when smoking status is not recorded in patients admitted as an emergency, who are also more likely to have worse outcomes from surgery.

Missing not at random data are important, can alter your conclusions, and are the most difficult to diagnose and handle. They can only be detected by collecting and examining some of the missing data. This is often difficult or impossible to do.

How you deal with missing data is dependent on the type of missingness. Once you know this, then you can sort it.

More on this below.

1. Ensure your data are coded correctly: ff_glimpse

While clearly obvious, this step is often ignored in the rush to get results. The first step in any analysis is robust data cleaning and coding. Lots of packages have a glimpse function and finalfit is no different. This function has three specific goals:

  1. Ensure all factors and numerics are correctly assigned. That is the commonest reason to get an error with a finalfit function. You think you’re using a factor variable, but in fact it is incorrectly coded as a continuous numeric.
  2. Ensure you know which variables have missing data. This presumes missing values are correctly assigned NA. See here for more details if you are unsure.
  3. Ensure factor levels and variable labels are assigned correctly.

Example scenario

Using the colon cancer dataset that comes with finalfit, we are interested in exploring the association between a cancer obstructing the bowel and 5-year survival, accounting for other patient and disease characteristics.

For demonstration purposes, we will create random MCAR and MAR smoking variables to the dataset.

The function summarises a data frame or tibble by numeric (continuous) variables and factor (discrete) variables. The dependent and explanatory  are for convenience. Pass either or neither e.g. to summarise data frame or tibble:

It doesn’t present well if you have factors with lots of levels, so you may want to remove these.

Use this to check that the variables are all assigned and behaving as expected. The proportion of missing data can be seen, e.g. smoking_mar has 23% missing data.

2. Identify missing values in each variable: missing_plot

In detecting patterns of missingness, this plot is useful. Row number is on the x-axis and all included variables are on the y-axis. Associations between missingness and observations can be easily seen, as can relationships of missingness between variables.

Click to enlarge.

It was only when writing this post that I discovered the amazing package, naniar. This package is recommended and provides lots of great visualisations for missing data.

3. Look for patterns of missingness: missing_pattern

missing_pattern simply wraps mice::md.pattern using finalfit grammar. This produces a table and a plot showing the pattern of missingness between variables.

This allows us to look for patterns of missingness between variables. There are 14 patterns in this data. The number and pattern of missingness help us to determine the likelihood of it being random rather than systematic. 

Make sure you include missing data in demographics tables

Table 1 in a healthcare study is often a demographics table of an “explanatory variable of interest” against other explanatory variables/confounders. Do not silently drop missing values in this table. It is easy to do this correctly with summary_factorlist. This function provides a useful summary of a dependent variable against explanatory variables. Despite its name, continuous variables are handled nicely.

na_include=TRUE ensures missing data from the explanatory variables (but not dependent) are included. Note that any p-values are generated across missing groups as well, so run a second time with na_include=FALSE if you wish a hypothesis test only over observed data.

4. Check for associations between missing and observed data: missing_pairs | missing_compare

In deciding whether data is MCAR or MAR, one approach is to explore patterns of missingness between levels of included variables. This is particularly important (I would say absolutely required) for a primary outcome measure / dependent variable.

Take for example “death”. When that outcome is missing it is often for a particular reason. For example, perhaps patients undergoing emergency surgery were less likely to have complete records compared with those undergoing planned surgery. And of course, death is more likely after emergency surgery.

missing_pairs uses functions from the excellent GGally package. It produces pairs plots to show relationships between missing values and observed values in all variables.

For continuous variables (age and nodes), the distributions of observed and missing data can be visually compared. Is there a difference between age and mortality above?

For discrete, data, counts are presented by default. It is often easier to compare proportions:

It should be obvious that missingness in Smoking (MCAR) does not relate to sex (row 6, column 3). But missingness  in Smoking (MAR) does differ by sex (last row, column 3) as was designed above when the missing data were created.

We can confirm this using missing_compare.

It takes “dependent” and “explanatory” variables, but in this context “dependent” just refers to the variable being tested for missingness against the “explanatory” variables.

Comparisons for continuous data use a Kruskal Wallis and for discrete data a chi-squared test.

As expected, a relationship is seen between Sex and Smoking (MAR) but not Smoking (MCAR).

For those who like an omnibus test

If you are work predominately with numeric rather than discrete data (categorical/factors), you may find these tests from the MissMech package useful. The package and output is well documented, and provides two tests which can be used to determine whether data are MCAR.

5. Decide how to handle missing data

These pages from Karen Grace-Martin are great for this.

Prior to a standard regression analysis, we can either:

  • Delete the variable with the missing data
  • Delete the cases with the missing data
  • Impute (fill in) the missing data
  • Model the missing data

MCAR, MAR, or MNAR

MCAR vs MAR

Using the examples, we identify that Smoking (MCAR) is missing completely at random. 

We know nothing about the missing values themselves, but we know of no plausible reason that the values of the missing data, for say, people who died should be different to the values of the missing data for those who survived. The pattern of missingness is therefore not felt to be MNAR.

Common solution

Depending on the number of data points that are missing, we may have sufficient power with complete cases to examine the relationships of interest.

We therefore elect to simply omit the patients in whom smoking is missing. This is known as list-wise deletion and will be performed by default in standard regression analyses including finalfit.

Other considerations

  1. Sensitivity analysis
  2. Omit the variable
  3. Imputation
  4. Model the missing data

If the variable in question is thought to be particularly important, you may wish to perform a sensitivity analysis. A sensitivity analysis in this context aims to capture the effect of uncertainty on the conclusions drawn from the model. Thus, you may choose to re-label all missing smoking values as “smoker”, and see if that changes the conclusions of your analysis. The same procedure can be performed labeling with “non-smoker”.

If smoking is not associated with the explanatory variable of interest (bowel obstruction) or the outcome, it may be considered not to be a confounder  and so could be omitted. That neatly deals with the missing data issue, but of course may not be appropriate.

Imputation and modelling are considered below.

MCAR vs MAR

But life is rarely that simple.

Consider that the smoking variable is more likely to be missing if the patient is female (missing_compareshows a relationship). But, say, that the missing values are not different from the observed values. Missingness is then MAR.

If we simply drop all the cases (patients) in which smoking is missing (list-wise deletion), then proportionality we drop more females than men. This may have consequences for our conclusions if sex is associated with our explanatory variable of interest or outcome.

Common solution

mice is our go to package for multiple imputation. That’s the process of filling in missing data using a best-estimate from all the other data that exists. When first encountered, this doesn’t sounds like a good idea.

However, taking our simple example, if missingness in smoking is predicted strongly by sex, and the values of the missing data are random, then we can impute (best-guess) the missing smoking values using sex and other variables in the dataset.

Imputation is not usually appropriate for the explanatory variable of interest or the outcome variable. With both of these, the hypothesis is that there is an meaningful association with other variables in the dataset, therefore it doesn’t make sense to use these variables to impute them.

Here is some code to run mice. The package is well documented, and there are a number of checks and considerations that should be made to inform the imputation process. Read the documentation carefully prior to doing this yourself.

The final table can easily be exported to Word or as a PDF as described else where.

By examining the coefficients, the effect of the imputation compared with the complete case analysis can be clearly seen.

Other considerations

  1. Omit the variable
  2. Imputing factors with new level for missing data
  3. Model the missing data

As above, if the variable does not appear to be important, it may be omitted from the analysis. A sensitivity analysis in this context is another form of imputation. But rather than using all other available information to best-guess the missing data, we simply assign the value as above. Imputation is therefore likely to be more appropriate.

There is an alternative method to model the missing data for the categorical in this setting – just consider the missing data as a factor level. This has the advantage of simplicity, with the disadvantage of increasing the number of terms in the model. Multiple imputation is generally preferred. 

MNAR vs MAR

Missing not at random data is tough in healthcare. To determine if data are MNAR for definite, we need to know their value in a subset of observations (patients).

Using our example above. Say smoking status is poorly recorded in patients admitted to hospital as an emergency with an obstructing cancer. Obstructing bowel cancers may be larger or their position may make the prognosis worse. Smoking may relate to the aggressiveness of the cancer and may be an independent predictor of prognosis. The missing values for smoking may therefore not random. Smoking may be more common in the emergency patients and may be more common in those that die.

There is no easy way to handle this. If at all possible, try to get the missing data. Otherwise, take care when drawing conclusions from analyses where data are thought to be missing not at random. 

Where to next

We are now doing more in Stan. Missing data can be imputed directly within a Stan model which feels neat. Stan doesn’t yet have the equivalent of NA which makes passing the data block into Stan a bit of a faff. 

Alternatively, the missing data can be directly modelled in Stan. Examples are provided in the manual. Again, I haven’t found this that easy to do, but there are a number of Stan developments that will hopefully make this more straightforward in the future. 

Finalfit now includes bootstrap simulation for model prediction

If your new to modelling in R and don’t know what this title means, you definitely want to look into doing it.

I’ve always been a fan of converting model outputs to real-life quantities of interest. For example, I like to supplement a logistic regression model table with predicted probabilities for a given set of explanatory variable levels. This can be more intuitive than odds ratios, particularly for a lay audience.

For example, say I have run a logistic regression model for predicted 5 year survival after colon cancer. What is the actual probability of death for a patient under 40 with a small cancer that has not perforated? How does that probability differ for a patient over 40?

I’ve tried this various ways. I used Zelig for a while including here, but it started trying to do too much and was always broken (I updated it the other day in the hope that things were better, but was met with a string of errors again).

I also used rms, including here (checkout the nice plots!). I like it and respect the package. But I don’t use it as standard and so need to convert all the models first, e.g. to lrm. Again, for my needs it tries to do too much and I find datadist awkward.

Thirdly, I love Stan for this, e.g. used in this paper. The generated quantities block allows great flexibility to simulate whatever you wish from the posterior. I’m a Bayesian at heart will always come back to this. But for some applications it’s a bit much, and takes some time to get running as I want.

I often simply want to predict y-hat from lm and glm with bootstrapped intervals and ideally a comparison of explanatory levels sets. Just like sim does in Zelig. But I want it in a format I can immediately use in a publication.

Well now I can with finalfit.

You need to use the github version of the package until CRAN is updated

There’s two main functions with some new internals to help expand to other models in the future.

Create new dataframe of explanatory variable levels

finalfit_newdata is used to generate a new dataframe. I usually want to set 4 or 5 combinations of x levels and often find it difficult to get this formatted for predict. Pass the original dataset, the names of explanatory variables used in the model, and a list of levels for these. For the latter, they can be included as rows or columns. If the data type is incorrect or you try to pass factor levels that don’t exist, it will fail with a useful warning.

Run bootstrap simulations of model predictions

boot_predict takes standard lm and glm model objects, together with finalfit lmlist and glmlist objects from fitters, e.g. lmmulti and glmmulti. In addition, it requires a newdata object generated from finalfit_newdata. If you’re new to this, don’t be put off by all those model acronyms, it is straightforward.

Note that the number of simulations (R) here is low for demonstration purposes. You should expect to use 1000 to 10000 to ensure you have stable estimates.

Output to Word, PDF, and html via RMarkdown

Simulations are produced using bootstrapping and everything is tidily outputted in a table/dataframe, which can be passed to knitr::kable.

Make comparisons

Better still, by including boot_compare==TRUE, comparisons are made between the first row of newdata and each subsequent row. These can be first differences (e.g. absolute risk differences) or ratios (e.g. relative risk ratios). The comparisons are done on the individual bootstrap predictions and the distribution summarised as a mean with percentile confidence intervals (95% CI as default, e.g. 2.5 and 97.5 percentiles). A p-value is generated on the proportion of values on the other side of the null from the mean, e.g. for a ratio greater than 1.0, p is the number of bootstrapped predictions under 1.0. Multiplied by two so it is two-sided. (Sorry about including a p-value).

Scroll right here:

What is not included?

It doesn’t yet include our other common models, such as coxph which I may add in. It doesn’t do lmer or glmer either. bootMer works well mixed-effects models which take a bit more care and thought, e.g. how are random effects to be handled in the simulations. So I don’t have immediate plans to add that in, better to do directly.

Plotting

Finally, as with all finalfit functions, results can be produced as individual variables using condense == FALSE. This is particularly useful for plotting

So there you have it. Straightforward bootstrapped simulations of model predictions, together with comparisons and easy plotting.

Finalfit now in CRAN

Your favourite package for getting model outputs directly into publication ready tables is now available on CRAN. They make you work for it! Thank you to all that helped. The development version will continue to be available from github.

Finalfit, knitr and R Markdown for quick results

Thank you for the many requests to provide some extra info on how best to get finalfit results out of RStudio, and particularly into Microsoft Word.

Here is how.

Make sure you are on the most up-to-date version of finalfit.

What follows is for demonstration purposes and is not meant to illustrate model building.

Does a tumour characteristic (differentiation) predict 5-year survival?

Demographics table

First explore variable of interest (exposure) by making it the dependent.

Note this useful alternative way of specifying explanatory variable lists:

Look at associations between our exposure and other explanatory variables. Include missing data.

Note missing data in obstruct.factor. We will drop this variable for now (again, this is for demonstration only). Also see that nodes has not been labelled.
There are small numbers in some variables generating chisq.test warnings (predicted less than 5 in any cell). Generate final table.

Logistic regression table

Now examine explanatory variables against outcome. Check plot runs ok.

Odds ratio plot

To MS Word via knitr/R Markdown

Important. In most R Markdown set-ups, environment objects require to be saved and loaded to R Markdown document.

We use RStudio Server Pro set-up on Ubuntu. But these instructions should work fine for most/all RStudio/Markdown default set-ups.

In RStudio, select File > New File > R Markdown.

A useful template file is produced by default. Try hitting knit to Word on the knitr button at the top of the .Rmd script window.

Now paste this into the file:

It’s ok, but not great.

Create Word template file

Now, edit the Word template. Click on a table. The style should be compact. Right click > Modify... > font size = 9. Alter heading and text styles in the same way as desired. Save this as template.docx. Upload to your project folder. Add this reference to the .Rmd YAML heading, as below. Make sure you get the space correct.

The plot also doesn’t look quite right and it prints with warning messages. Experiment with fig.width to get it looking right.

Now paste this into your .Rmd file and run:

This is now looking good for me, and further tweaks can be made.

To PDF via knitr/R Markdown

Default settings for PDF:

Again, ok but not great.

We can fix the plot in exactly the same way. But the table is off the side of the page. For this we use the kableExtra package. Install this in the normal manner. You may also want to alter the margins of your page using geometry in the preamble.

This is now looking pretty good for me as well.

There you have it. A pretty quick workflow to get final results into Word and a PDF.

Elegant regression results tables and plots in R: the finalfit package

The finafit package brings together the day-to-day functions we use to generate final results tables and plots when modelling. I spent many years repeatedly manually copying results from R analyses and built these functions to automate our standard healthcare data workflow. It is particularly useful when undertaking a large study involving multiple different regression analyses. When combined with RMarkdown, the reporting becomes entirely automated. Its design follows Hadley Wickham’s tidy tool manifesto.

Installation and Documentation

It lives on GitHub.

You can install finalfit from github with:

It is recommended that this package is used together with dplyr, which is a dependent.

Some of the functions require rstan and boot. These have been left as Suggests rather than Depends to avoid unnecessary installation. If needed, they can be installed in the normal way:

To install off-line (or in a Safe Haven), download the zip file and use devtools::install_local().

Main Features

1. Summarise variables/factors by a categorical variable

summary_factorlist() is a wrapper used to aggregate any number of explanatory variables by a single variable of interest. This is often “Table 1” of a published study. When categorical, the variable of interest can have a maximum of five levels. It uses Hmisc::summary.formula().

See other options relating to inclusion of missing data, mean vs. median for continuous variables, column vs. row proportions, include a total column etc.

summary_factorlist() is also commonly used to summarise any number of variables by an outcome variable (say dead yes/no).

Tables can be knitted to PDF, Word or html documents. We do this in RStudio from a .Rmd document. Example chunk:

2. Summarise regression model results in final table format

The second main feature is the ability to create final tables for linear (lm()), logistic (glm()), hierarchical logistic (lme4::glmer()) and
Cox proportional hazards (survival::coxph()) regression models.

The finalfit() “all-in-one” function takes a single dependent variable with a vector of explanatory variable names (continuous or categorical variables) to produce a final table for publication including summary statistics, univariable and multivariable regression analyses. The first columns are those produced by summary_factorist(). The appropriate regression model is chosen on the basis of the dependent variable type and other arguments passed.

Logistic regression: glm()

Of the form: glm(depdendent ~ explanatory, family="binomial")

Logistic regression with reduced model: glm()

Where a multivariable model contains a subset of the variables included specified in the full univariable set, this can be specified.

Mixed effects logistic regression: lme4::glmer()

Of the form: lme4::glmer(dependent ~ explanatory + (1 | random_effect), family="binomial")

Hierarchical/mixed effects/multilevel logistic regression models can be specified using the argument random_effect. At the moment it is just set up for random intercepts (i.e. (1 | random_effect), but in the future I’ll adjust this to accommodate random gradients if needed (i.e. (variable1 | variable2).

Cox proportional hazards: survival::coxph()

Of the form: survival::coxph(dependent ~ explanatory)

Add common model metrics to output

metrics=TRUE provides common model metrics. The output is a list of two dataframes. Note chunk specification for output below.

Rather than going all-in-one, any number of subset models can be manually added on to a summary_factorlist() table using finalfit_merge(). This is particularly useful when models take a long-time to run or are complicated.

Note the requirement for fit_id=TRUE in summary_factorlist(). fit2df extracts, condenses, and add metrics to supported models.

Bayesian logistic regression: with stan

Our own particular rstan models are supported and will be documented in the future. Broadly, if you are running (hierarchical) logistic regression models in [Stan](http://mc-stan.org/users/interfaces/rstan) with coefficients specified as a vector labelled beta, then fit2df() will work directly on the stanfit object in a similar manner to if it was a glm or glmerMod object.

3. Summarise regression model results in plot

Models can be summarized with odds ratio/hazard ratio plots using or_plot, hr_plot and surv_plot.

OR plot

HR plot

Kaplan-Meier survival plots

KM plots can be produced using the library(survminer)

Notes

Use Hmisc::label() to assign labels to variables for tables and plots.

Export dataframe tables directly or to R Markdown knitr::kable().

Note wrapper summary_missing() is also useful. Wraps mice::md.pattern.

Development will be on-going, but any input appreciated.

Install github package on safe haven server

I’ve had few enquires about how to install the summarizer package on a server without internet access, such as the NHS Safe Havens.

  1. Uploadsummarizer-master.zip from here to server.
  2. Unzip.
  3. Run this:


library(devtools)
source = devtools:::source_pkg("summarizer-master")
install(source)

Edit

As per comments, devtools::install_local() has previously failed, but may now also work directly.

P-values from random effects linear regression models

lme4::lmer is a useful frequentist approach to hierarchical/multilevel linear regression modelling. For good reason, the model output only includes t-values and doesn’t include p-values (partly due to the difficulty in estimating the degrees of freedom, as discussed here).

Yes, p-values are evil and we should continue to try and expunge them from our analyses. But I keep getting asked about this. So here is a simple bootstrap method to generate two-sided parametric p-values on the fixed effects coefficients. Interpret with caution.

 

An alternative presentation of the ProPublica Surgeon Scorecard

ProPublica, an independent investigative journalism organisation, have published surgeon-level complications rates based on Medicare data. I have already highlighted problems with the reporting of the data: surgeons are described as having a “high adjusted rate of complications” if they fall in the red-zone, despite there being too little data to say whether this has happened by chance.

4
This surgeon should not be identified as having a “high adjusted rate of complications” as there are too few cases to estimate the complication rate accurately.

I say again, I fully support transparency and public access to healthcare. But the ProPublica reporting has been quite shocking. I’m not aware of them publishing the number of surgeons out of the 17000 that are statistically different to the average. This is a small handful.

ProPublica could have chosen a different approach. This is a funnel plot and I’ve written about them before.

A funnel plot is a summary of an estimate (such as complication rate) against a measure of the precision of that estimate. In the context of healthcare, a centre or individual outcome is often plotted against patient volume. A horizontal line parallel to the x-axis represents the outcome for the entire population and outcomes for individual surgeons are displayed as points around this. This allows a comparison of individuals with that of the population average, while accounting for the increasing certainty surrounding that outcome as the sample size increases. Limits can be determined, beyond which the chances of getting an individual outcome are low if that individual were really part of the whole population.

In other words, a surgeon above the line has a complication rate different to the average.

I’ve scraped the ProPublica data for gallbladder removal (laparoscopic cholecystectomy) from California, New York and Texas for surgeons highlighted in the red-zone. These are surgeons ProPublica says have high complication rates.

As can be seen from the funnel plot, these surgeons are no where near being outliers. There is insufficient information to say whether any of them are different to average. ProPublica decided to ignore the imprecision with which the complication rates are determined. For red-zone surgeons from these 3 states, none of them have complication rates different to average.

ProPublica_lap_chole_funnel
Black line, population average (4.4%), blue line 95% control limit, red line 99% control limit.

How likely is it that a surgeon with an average complication rate (4.4%) will appear in the red-zone just by chance (>5.2%)? The answer is, pretty likely given the small numbers of cases here: anything up to a 25% chance depending on the number of cases performed. Even at the top of the green-zone (low ACR, 3.9%), there is still around a 1 in 6 chance a surgeon will appear to have a high complication rate just by chance.

chance_of_being_in_redzoneProPublica have failed in their duty to explain these data in a way that can be understood. The surgeon score card should be revised. All “warning explanation points” should be removed for those other than the truly outlying cases.

Data

Download

Git

Link to repository.

Code

RStudio and GitHub

Version control has become essential for me keeping track of projects, as well as collaborating. It allows backup of scripts and easy collaboration on complex projects. RStudio works really well with Git, an open source open source distributed version control system, and GitHub, a web-based Git repository hosting service. I was always forget how to set up a repository, so here’s a reminder.

This example is done on RStudio Server, but the same procedure can be used for RStudio desktop. Git or similar needs to be installed first, which is straight forward to do.

Setup Git on RStudio and Associate with GitHub

In RStudio, Tools -> Version Control, select Git.

In RStudio, Tools -> Global Options, select Git//SVN tab. Ensure the path to the Git executable is correct. This is particularly important in Windows where it may not default correctly (e.g. C:/Program Files (x86)/Git/bin/git.exe).
1Now hit, Create RSA Key …

2_rsaClose this window.

Click, View public key, and copy the displayed public key.

4_rsaIf you haven’t already, create a GitHub account. Open your account settings and click the SSH keys tab. Click Add SSH key. Paste in the public key you have copied from RStudio.

6_add_keyTell Git who you are. Remember Git is a piece of software running on your own computer. This is distinct to GitHub, which is the repository website. In RStudio, click Tools -> Shell … . Enter:

git config --global user.email "mail@ewenharrison.com"
git config --global user.name "ewenharrison"

Use your GitHub username.

10_who_are_you

Create New project AND git

In RStudio, click New project as normal. Click New Directory.

7_new_project

Name the project and check Create a git repository.

8_new_project_with_git

Now in RStudio, create a new script which you will add to your repository.

9_test_scriptAfter saving your new script (test.R), it should appear in the Git tab on the Environment / history panel.

11_initial_commitClick the file you wish to add, and the status should turn to a green ‘A’. Now click Commit and enter an identifying message in Commit message.

12_inital_commit2You have now committed the current version of this file to your repository on your computer/server. In the future you may wish to create branches to organise your work and help when collaborating.

Now you want to push the contents of this commit to GitHub, so it is also backed-up off site and available to collaborators. In GitHub, create a New repository, called here test.

5_create_git In RStudio, again click Tools -> Shell … . Enter:

git remote add origin https://github.com/ewenharrison/test.git
git config remote.origin.url git@github.com:ewenharrison/test.git
git pull -u origin master
git push -u origin master

13_push_pullYou have now pushed your commit to GitHub, and should be able to see your files in your GitHub account. The Pull Push buttons in RStudio will now also work. Remember, after each Commit, you have to Push to GitHub, this doesn’t happen automatically.

Clone an existing GitHub project to new RStudio project

In RStudio, click New project as normal. Click Version Control.

7_new_projectIn Clone Git Repository, enter the GitHub repository URL as per below. Change the project directory name if necessary.

14_new_version_controlIn RStudio, again click Tools -> Shell … . Enter:

git config remote.origin.url git@github.com:ewenharrison/test.git

Interested in international trials? Take part in GlobalSurg.

Bayesian statistics and clinical trial conclusions: Why the OPTIMSE study should be considered positive

Statistical approaches to randomised controlled trial analysis

The statistical approach used in the design and analysis of the vast majority of clinical studies is often referred to as classical or frequentist. Conclusions are made on the results of hypothesis tests with generation of p-values and confidence intervals, and require that the correct conclusion be drawn with a high probability among a notional set of repetitions of the trial.

Bayesian inference is an alternative, which treats conclusions probabilistically and provides a different framework for thinking about trial design and conclusions. There are many differences between the two, but for this discussion there are two obvious distinctions with the Bayesian approach. The first is that prior knowledge can be accounted for to a greater or lesser extent, something life scientists sometimes have difficulty reconciling. Secondly, the conclusions of a Bayesian analysis often focus on the decision that requires to be made, e.g. should this new treatment be used or not.

There are pros and cons to both sides, nicely discussed here, but I would argue that the results of frequentist analyses are too often accepted with insufficient criticism. Here’s a good example.

OPTIMSE: Optimisation of Cardiovascular Management to Improve Surgical Outcome

Optimising the amount of blood being pumped out of the heart during surgery may improve patient outcomes. By specifically measuring cardiac output in the operating theatre and using it to guide intravenous fluid administration and the use of drugs acting on the circulation, the amount of oxygen that is delivered to tissues can be increased.

It sounds like common sense that this would be a good thing, but drugs can have negative effects, as can giving too much intravenous fluid. There are also costs involved, is the effort worth it? Small trials have suggested that cardiac output-guided therapy may have benefits, but the conclusion of a large Cochrane review was that the results remain uncertain.

A well designed and run multi-centre randomised controlled trial was performed to try and determine if this intervention was of benefit (OPTIMSE: Optimisation of Cardiovascular Management to Improve Surgical Outcome).

Patients were randomised to a cardiac output–guided hemodynamic therapy algorithm for intravenous fluid and a drug to increase heart muscle contraction (the inotrope, dopexamine) during and 6 hours following surgery (intervention group) or to usual care (control group).

The primary outcome measure was the relative risk (RR) of a composite of 30-day moderate or major complications and mortality.

OPTIMSE: reported results

Focusing on the primary outcome measure, there were 158/364 (43.3%) and 134/366 (36.6%) patients with complication/mortality in the control and intervention group respectively. Numerically at least, the results appear better in the intervention group compared with controls.

Using the standard statistical approach, the relative risk (95% confidence interval) = 0.84 (0.70-1.01), p=0.07 and absolute risk difference = 6.8% (−0.3% to 13.9%), p=0.07. This is interpreted as there being insufficient evidence that the relative risk for complication/death is different to 1.0 (all analyses replicated below). The authors reasonably concluded that:

In a randomized trial of high-risk patients undergoing major gastrointestinal surgery, use of a cardiac output–guided hemodynamic therapy algorithm compared with usual care did not reduce a composite outcome of complications and 30-day mortality.

A difference does exist between the groups, but is not judged to be a sufficient difference using this conventional approach.

OPTIMSE: Bayesian analysis

Repeating the same analysis using Bayesian inference provides an alternative way to think about this result. What are the chances the two groups actually do have different results? What are the chances that the two groups have clinically meaningful differences in results? What proportion of patients stand to benefit from the new intervention compared with usual care?

With regard to prior knowledge, this analysis will not presume any prior information. This makes the point that prior information is not always necessary to draw a robust conclusion. It may be very reasonable to use results from pre-existing meta-analyses to specify a weak prior, but this has not been done here. Very grateful to John Kruschke for the excellent scripts and book, Doing Bayesian Data Analysis.

The results of the analysis are presented in the graph below. The top panel is the prior distribution. All proportions for the composite outcome in both the control and intervention group are treated as equally likely.

The middle panel contains the main findings. This is the posterior distribution generated in the analysis for the relative risk of the composite primary outcome (technical details in script below).

The mean relative risk = 0.84 which as expected is the same as the frequentist analysis above. Rather than confidence intervals, in Bayesian statistics a credible interval or region is quoted (HDI = highest density interval is the same). This is philosphically different to a confidence interval and says:

Given the observed data, there is a 95% probability that the true RR falls within this credible interval.

This is a subtle distinction to the frequentist interpretation of a confidence interval:

Were I to repeat this trial multiple times and compute confidence intervals, there is a 95% probability that the true RR would fall within these confidence intervals.

This is an important distinction and can be extended to make useful probabilistic statements about the result.

The figures in green give us the proportion of the distribution above and below 1.0. We can therefore say:

The probability that the intervention group has a lower incidence of the composite endpoint is 97.3%.

It may be useful to be more specific about the size of difference between the control and treatment group that would be considered equivalent, e.g. 10% above and below a relative risk = 1.0. This is sometimes called the region of practical equivalence (ROPE; red text on plots). Experts would determine what was considered equivalent based on many factors. We could therefore say:

The probability of the composite end-point for the control and intervention group being equivalent is 22%.

Or, the probability of a clinically relevant difference existing in the composite endpoint between control and intervention groups is 78%

optimise_primary_bayesFinally, we can use the 200 000 estimates of the probability of complication/death in the control and intervention groups that were generated in the analysis (posterior prediction). In essence, we can act like these are 2 x 200 000 patients. For each “patient pair”, we can use their probability estimates and perform a random draw to simulate the occurrence of complication/death. It may be useful then to look at the proportion of “patients pairs” where the intervention patient didn’t have a complication but the control patient did:

Using posterior prediction on the generated Bayesian model, the probability that a patient in the intervention group did not have a complication/death when a patient in the control group did have a complication/death is 28%.

Conclusion

On the basis of a standard statistical analysis, the OPTIMISE trial authors reasonably concluded that the use of the intervention compared with usual care did not reduce a composite outcome of complications and 30-day mortality.

Using a Bayesian approach, it could be concluded with 97.3% certainty that use of the intervention compared with usual care reduces the composite outcome of complications and 30-day mortality; that with 78% certainty, this reduction is clinically significant; and that in 28% of patients where the intervention is used rather than usual care, complication or death may be avoided.

A map of the world by tweets

With geo-tagging enabled, tweets include information on the location of the user when the tweet was sent. Miguel Rios (@miguelrios) has plotted locations of billions of tweets to create maps of the world. This is pretty amazing stuff – a world map rendered just from twitter posts!

Maps are created using every tweet from 2009 using R and the ggmap package.

 

Europe by tweets,  by Miguel Rios @miguelrios
Europe by tweets, by Miguel Rios @miguelrios

 

US by tweets,  by Miguel Rios @miguelrios
US by tweets, by Miguel Rios @miguelrios

 

Post is here with more here on flickr.

Mortality after paediatric heart surgery using public domain data

This post comes with some big health warnings.

The recent events in Leeds highlight the difficulties faced in judging the results of surgery by individual hospital. A clear requirement is timely access to data in a form easily digestible by the public.

Here I’ve scraped the publically available data from the central cardiac audit database (CCAD). All the data are available at the links provided and are as presented this afternoon (06/04/2013). Please read the caveats carefully.

Hospital-specific 30-day mortality data are available for certain paediatric heart surgery procedures for 2009-2012. These data are not complete for 2011-12 and there may be missing data for earlier years. There may be important data for procedures not included here that should be accounted for. There is no case-mix adjustment.

All data are included in spreadsheets below as well as the code to run the analysis yourself, to ensure no mistakes have been made. Hopefully these data will be quickly superseded with a quality-assured update.

Mortality by centre

The funnel plot below has been generated by taking all surgical procedures performed from pages such as this and expressing all deaths within 30 days as a proportion of the total procedures performed by hospital.

The red horizontal line is the mean mortality rate for these procedures – 2.3%. The green, blue and red curved lines are decreasingly stringent control limits within which unit results may vary by chance.

ccad_funnel_april_2013

Mortality by procedure

The mortality associated with different procedures can be explored with this google motion chart. Note when a procedure is uncommon (to the left of the chart) the great variation seen year to year. These bouncing balls trace out the limits of a funnel plot. They highlight why year to year differences in mortality rates for rare procedures must be interpreted with caution.

 

Data

ccad_public_data_april_2013.xls

ccad_public_data_april_2013_centre

ccad_public_data_april_2013_aggregate

ccad_public_data_april_2013_lookup

 

Script