What causes a stroke and why does it happen? [Infographic]


Before answering the above question, let’s begin with understanding.
What is a stroke?
A stroke is a damage to the brain, that occurs when the blood supply to part of your brain is interrupted or reduced, preventing brain tissue from getting oxygen and nutrients. Brain cells begin to die in minutes. This causes the part of the body that the injured brain controls to stop working. A stroke also is called a cerebrovascular accident, CVA, or “brain attack.”

Figure 0: Brain Stroke

There are many myths regarding strokes which are as follows:

Strokes only happen to elderly people.
Strokes are rare.
Smoking doesn’t affect your chances of having a stroke.
Males are most susceptible to strokes.
Being Married will increase the risk of having a stroke.
And many more…

Therefore, I used data from Kaggle, a Stroke Prediction Dataset that consists of 11 clinical features for predicting stroke events and to break down some myths using the features.

This dataset contains many clinical features like Age, Heart Diseases (if any), BMI (body mass index), sugar level, previously affected cases, etc. So, from this dataset, we can predict how likely a person can get a stroke based on his remaining features, and also, we can break some myths which include:

1.Strokes only happen to elderly people.

2.Smoking doesn’t affect your chances of having a stroke.

3.Blood sugar and BMI do not contribute to stroke.

Myth I: Strokes only happen to elderly people.

Fact: It is true that as you age your risk for stroke goes up. However, there’s also an increasing number of strokes in people between the ages of 18 and 65, so to say that strokes only occur in the elderly is false. The growing incidence of obesity and high blood pressure in ages 18 to 65 may contribute to the increased stroke risk in this population.

Now let us see what our data-based approach suggests:
In our dataset, there are a total of 5,110 rows and 12 columns, so there are a total of 5,110 ages and stroke/no-stroke cases recorded in our dataset. In our dataset, the number of people who don’t have a stroke is higher than the number of people who have a stroke, which we refer to as an unbalanced dataset.

Figure1: Count Plot on Stroke column

Now we have plotted a KDE plot to know the impact of age on the stroke column in our unbalanced dataset and we can see that stroke in people started from age 30 and continues to age 90.

Figure2: Age vs Stroke (Unbalanced)

Let’s make our dataset balanced and recheck our previous results. An Unbalanced dataset can be handled using many techniques, but, in our case, we used Random Oversampling and after oversampling our data here is a picture of a balanced stroke column.

Figure 3: Balanced Count plot on Stroke Column
Figure 4: Age vs Stroke (Balanced)

It is now clear that not only older people but also people over the age of 20 are at the same risk of having a stroke.

Myth II: Smoking doesn’t affect your chances of having a stroke.

Fact: Smoking is one of the biggest risk factors for stroke, especially in young people. This is true for both ischemic and hemorrhagic strokes, as well as first-time and recurrent strokes.
Ischemic stroke: This kind of stroke is caused by a blockage in an artery that provides blood to the brain. Blockage decreases blood flow and oxygen to the brain, causing damage or death of brain cells.
Hemorrhagic stroke: A hemorrhagic stroke is also called an intracerebral haemorrhage, or ICH. ICH occurs when a blood vessel ruptures and blood accumulates in the tissue around the rupture. This puts pressure on the brain and causes blood loss in the surroundings.

Let’s take a look at what our data-driven approach suggests:
According to our analysis, there is evidence that, regardless of the person’s current smoking status, if the person smokes, the chances of having a stroke increase.

Figure 5: Smoking Status vs Stroke

As we can see here, the image suggests that those who have already smoked and those who currently smoke have a higher percentage of strokes, compared to people who never smoked.

It breaks the myth and proves that smoking is one of the most important risk factors for stroke.

Myth III: Blood sugar and BMI do not contribute to stroke.

Fact: Diabetes is a well-established risk factor for stroke. It can cause pathological changes in blood vessels at various locations and can lead to a stroke if the cerebral vessels are directly affected. Additionally, mortality is higher and poststroke outcomes are poorer in patients with uncontrolled glucose levels. Another effect of being overweight is that the body’s metabolism changes in ways that lead to an excess of circulating lipids, high cholesterol, and elevated blood glucose, all of which, over time, harm the blood vessels of the brain and the heart and lead to the formation of stroke-causing blood clots in the heart and brain.

Let’s now turn to what our evidence-based approach suggests:
We first analyzed the BMI column in our dataset and found that people who are overweight are more susceptible to stroke.

Figure 6: BMI vs stroke

As we can see, people with a BMI of more than 28 are the most at risk of having a stroke. The mean glucose levels are thus as follows.

Figure 7: Average Glucose level vs stroke

The image shows that people with low glucose levels as well as people with higher glucose levels are at equal risk of having a stroke. Hypoglycemia (lower blood sugar) is defined as blood sugar under 70 mg/dl. This is common in diabetic patients who are treated pharmacologically. Transient hypoglycemia is well known to produce a stroke-like picture with hemiplegia and aphasia. The effect of being overweight is that the body’s metabolism changes in ways that lead to an excess of circulating lipids, high cholesterol, and elevated blood glucose. As a result of this elevated blood sugar level, the risk of stroke increases exponentially.

Thus, it breaks the myth and proves that glucose levels and BMI play a major role in the onset of a stroke.

Stroke Prevention
1.Proper Diet
3.Stop Smoking
4.Cut down on Alcohol
The lifestyle changes mentioned above can help prevent a stroke.

A stroke is a medical emergency, and prompt treatment is crucial. Early action can reduce brain damage and other complications.


In this article, we examined the causes of a stroke, using Kaggle’s 2021 health care data.

  • First, we broke a myth that says “Strokes only happen to elderly people”. It is true that with age, the risk of stroke increases. However, there is also a growing number of strokes among individuals between the ages of 20 and 40.
  • Second, we broke another myth that says “Smoking doesn’t affect your chances of having a stroke”. We demonstrated that Smoking is one of the greatest hazard components for stroke, particularly in youthful individuals.
  • Finally, we proved that Glucose level and BMI both together, play an imperative part in raising the chance of having a stroke.

So, this is my analysis of how and why a brain stroke can occur. But the real question is:

Are YOU healthy enough for not having a stroke?

To see more about this analysis, see the link to my GitHub available here.




love to analyze things.

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

How SnackNation Leverages Machine Learning to Predict (and Prevent) Customer Churn

How to Run an Exceptional Data Science Team

Big Data Analytics: Get an In-depth Look

Drones Raise the Bar for Roadway Pavement Inspection

How to Deliver on Your Data Strategy

Causal Inference cheat sheet for data scientists

(The challenges of) Predicting cardiovascular disease from Electronic Health Records using…

Classification Model for Loan Default Risk Prediction

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Lakshman Raj

Lakshman Raj

love to analyze things.

More from Medium

Stop with the pictures

Data-Driven Hiring

Klaviyo Data Science Podcast EP 19 | 2021: a data science year in review

An Exploration of Seattle Airbnb Data