Tutorial 3

Categories → Tutorial

Does quitting smoking cause weight gain?

This example is reproduced from Chapter 12 of Causal Inference: What If (the book), by Miguel Hernan. The book is a great and thorough course on Causal Inference, and highly recommended for those who want to get deeply into it.

This case study uses data from the National Health and Nutrition Examination Survey Data Epidemiologic Follow-up Study (NHEFS). It attempts to accurately quantify the effect of smoking cessation on weight gain. In short, do people gain weight after giving up smoking?

We can see that despite some simplifications to make this tutorial easy, we obtain a result close to the reported value given in the book.

Step 1: Create a new Study

1. Click on Studies in the menu bar.
2. In the Studies list page, click the Create button.
3. Give the new Study any name you like (e.g. Tutorial 3).
4. Select the Dataset with the name beginning "Tutorial 3"
5. Click Create and the new Study will be displayed.

Step 2: Specify Treatment and Outcome

Select variable Quit Smoking? as the Treatment.

Select Change in Weight as the Outcome.

Causal Wizard automatically detects the data type in the Treatment column and modifies the user interface to match.

Since quit smoking is provided in the data as a binary value, we don't see the threshold tool this time. Instead, we are asked to specify the value for Control (didn't quit smoking) and Treated (did quit). The defaults are 0 and 1 respectively, which is fine.

Step 3: Draw Causal Diagram

This case study uses all the numerical variables in the dataset. Create the graph as shown in the picture below (note: nodes are automatically coloured during the check process; verify that the colours are the same after the Check step to be sure you've added all the same edges). 

Basically, every variable is treated as a "confounder" - a cause of both Treatment and Outcome.

In addition, there's one edge directly from Treatment Quit Smoking?  to Outcome Change in Weight.

Step 4: Check the setup

1. Press the Check button at the top of the page. 
2. If something is not right, you'll get some warning messages - follow their instructions to complete or fix the setup.
3. If everything went as planned, you'll see a large green block which invites you to select a Model type. Select "Backdoor: Propensity Score-based Inverse Weighting". This model is chosen because it was used in the original example in Miguel's book.
4. Press the Calculate button below the green block.
5. Click confirm.

Step 5: Reviewing the results

You can always come back to earlier results. If you decided to come back later while following this tutorial, you can find your result by:

1. Clicking on Studies in the menu bar.
2. In the Studies list page, click the Results button next to your Tutorial 3 Study.
3. You'll now see all Results for this Study. If you've followed this tutorial, you'll only have one. Click the Open button.

Once opened, look at the first row of the findings, which should look something like this:

• The Average Treatment Effect (ATE) of treatment Quit Smoking on outcome Change in Weight was estimated as 3.05. (95% Confidence Interval [2.11, 3.85] )

Note that on Chapter 12, page 158 of "What If" an ATE of 3.4 kg is obtained. This is well within the 95% confidence interval of our simplified model given above. 

Another thing worth noting is that the placebo treatment validation test failed - perhaps due to dataset size and noise, but also perhaps due to simplified modelling. These validation tests are quite strict - note that the placebo treatment effect is much smaller, and quite close to zero even though not quite zero.