Part 4 – how to work out whether snus is bad for your health.

So far in this series on snus (my Master’s thesis topic) we’ve covered what snus is, the history behind it and some of its health effects using the scientific evidence available. But how do scientific researchers and public health experts actually work out the conclusions they come to? In this part we’ll use my snus thesis topic as an example to explain the how public health and epidemiological* research is conducted.

* the study of disease patterns, causes and effects of health in populations.

Step 1 – Become an expert in the field.

You really have to know your area of expertise before you can even start to think about researching it from scratch. Understanding what is already known about something is key to formulating a relevant and novel research question.

For example with snus, I performed a literature search using PubMed to understand the quality of evidence available on the subject. I was mainly concentrating on oral cancer as my research group had told me that project was available. After reading around 20-25 papers on the subject I started to get a feel for what was known and why my supervisors were interested in looking clearing up the evidence around snus and oral cancer.

TIP: Use Mendeley to keep track of your references and research papers! It’ll be handy for writing/citing your thesis and further projects if needed.

Step 2 – Formulate a research question.

Now you know what is missing in the literature, develop a rational question you that will add something novel to the pile of evidence.

EXAMPLE: You put snus in your mouth. Snus contains tobacco. Tobacco contains carcinogens. Carcinogens are in contact with your mouth. Can you get mouth cancer from snus?

Seems rational right? But what does the evidence say?

Well snus as we’ve seen in Part 3 – “the evidence regarding oral (mouth) cancer is debatable. Current research available is limited by small studies, poor design and difficulty adjusting for smoking and alcohol – which are strong risk factors for oral cancer.”

Basically some papers say YES it gives you mouth cancer, others say NO it doesn’t give you mouth cancer.

My research question:

“Is there any association between Swedish moist snuff (snus) use, cigarette smoking and risk of oral cancer?”

Step 3 – Find half a million people, ask them if they use snus and see if they get oral cancer in the next 30 years.

Easy huh? Fortunately, in Sweden this is quite easy… and a lot of the time its already done for you!

Using the Swedish personal number (personnummer) which is linked to hospital records, cancer databases, bank cards, your address, telephone number, tax and employment history, we can use surveys, studies and any other data collection method to “recruit” participants.

For example, my snus study is a mixture of 9 different studies which were started up to 30 years ago. At the beginning of these studies the participants were asked lifestyle questions like do you smoke? Do you snus? Do you drink alcohol? And often these were in quite a lot of detail regarding quantity and duration.

This is all recorded in a database and then linked to the national cancer registry. TA-DA! Alakazam! Wow! We suddenly have half a million people’s data, whether they use snus or smoke and if they had got oral cancer. Magic!

Step 4 – Use a computer, write some code and do some statistics.

Now we essentially have a huge database of people who were either exposed or not exposed at the beginning of them entering the study. This is the basis of a COHORT study. We then follow them over the study period and see if they develop the disease.

EXPOSED: Snus use or smoking

UNEXPOSED: Never used snus or never smoked

WITH OUTCOME: Gets oral cancer

WITHOUT OUTCOME: Doesn’t get oral cancer


By comparing the proportions of people in each exposure group who develop and don’t develop the disease. We can calculate something called the “relative risk” or “hazard ratio”. In my thesis, because we have a timescale to the study we are using Cox regression and are calculating* ‘hazard ratios’.

*my virtual friend STATA14 does the calculations for me

Also my friend (STATA) kindly works out whether this is statistically significant – for example has this happened by chance? Or is there something else going on – like snus or smoking causing more people to develop oral cancer.

Step 5 – Write a thesis, defend a thesis, publish a paper.

Dissemination of scientific research is probably the most important step. A professor once told me “You haven’t done it if you haven’t published it!” Once you get it out into the research community people will start talking about it, discussing it, challenging it and hopefully accepting it as a genuinely sound piece of research.

Or they’ll just tear it to shreds and ruin your academic career for ever….but that’s science hey!

Read more

This post is part of my series on snus and oral cancer – the topic of my master’s thesis. Read the three previous posts here:

Part 1 – What the deuce is snus?

Part 2 – The history of snus.

Part 3 – Is snus healthier than smoking?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s