An interactive Tableau dashboard that explores cardiovascular disease risk using demographic, lifestyle, and clinical patient data. This project demonstrates how Tableau can be used to transform raw healthcare records into meaningful visual insights that support data-driven decision-making.
This beginner Tableau project analyses 70,000 patient records to identify relationships between age, body mass index (BMI), lifestyle behaviours, blood pressure, and cardiovascular disease.
The dashboard combines multiple visualisations into a single interactive report, allowing users to filter the data and explore patterns associated with cardiovascular disease prevalence.
During the project, data cleaning, calculated fields, semantic modelling, and dashboard design best practices were applied to create a professional portfolio-ready solution.
- Explore cardiovascular disease prevalence across different patient groups.
- Analyse how age and BMI relate to disease risk.
- Investigate lifestyle factors associated with cardiovascular disease.
- Visualise the relationship between systolic and diastolic blood pressure.
- Build an interactive dashboard using Tableau best practices.
- Tableau Desktop
- Microsoft Excel
- Calculated Fields
- Dashboard Containers
- Interactive Filters
- Reference Lines
- Data Cleaning & Validation
- Source: Cardiovascular Disease Dataset (Kaggle)
- Records: 70,000 Patients
- Fields: 13
- Domain: Healthcare
Key variables include:
- Patient ID
- Age
- Gender
- Height
- Weight
- BMI (Calculated)
- Blood Pressure (Systolic & Diastolic)
- Cholesterol Level
- Glucose Level
- Smoking Status
- Alcohol Consumption
- Physical Activity
- Cardiovascular Disease Status
The dataset was prepared before visualisation by:
- Renaming columns to meaningful business names.
- Converting patient age from days to years.
- Creating Age Group classifications.
- Creating BMI calculations.
- Creating BMI Categories.
- Mapping gender values to Men/Women.
- Mapping cardiovascular disease values to Disease Present / No Disease.
- Validating boolean fields.
- Removing implausible blood pressure values for the scatter plot analysis.
- Building reusable calculated fields for dashboard metrics.
- Total Patients
- Disease Rate
- Average BMI
- Average Age
- Patient Age Distribution (Histogram)
- Disease Prevalence by Age
- Disease Prevalence by BMI
- Patient Lifestyle Risk Factors
- Blood Pressure Relationship (Scatter Plot)
- Gender
- Age Group
- BMI Category
- Cardiovascular Disease Status
Disease prevalence steadily increases across each age group.
| Age Group | Disease Present |
|---|---|
| Under 40 | 23.5% |
| 40–44 | 30.5% |
| 45–49 | 42.6% |
| 50–54 | 46.1% |
| 55–59 | 56.2% |
| 60 and Over | 66.7% |
Older patients exhibit substantially higher rates of cardiovascular disease, highlighting age as one of the strongest predictors in the dataset.
Disease prevalence rises consistently across BMI categories.
| BMI Category | Disease Present |
|---|---|
| Underweight | 27.5% |
| Healthy | 39.8% |
| Overweight | 53.6% |
| Obese | 67.6% |
Patients classified as obese show the highest disease prevalence, reinforcing the relationship between excess body weight and cardiovascular risk.
The patient population is concentrated between approximately 50 and 60 years old, with relatively few patients under the age of 40.
This demographic profile helps explain the overall disease rate observed across the dataset.
The scatter plot demonstrates a clear positive correlation between systolic and diastolic blood pressure.
Reference lines at:
- 140 mmHg (Systolic)
- 90 mmHg (Diastolic)
highlight recognised hypertension thresholds and show that many patients with cardiovascular disease cluster around elevated blood pressure values.
Across the patient population:
- Approximately 80% are physically active.
- Around 9% are smokers.
- Around 5% report alcohol consumption.
Although physical activity is common, cardiovascular disease remains prevalent among older and higher BMI groups, suggesting that age and obesity have stronger associations with disease prevalence within this dataset.
During exploratory analysis, implausible blood pressure values were identified, including:
- Negative blood pressure values
- Extremely high systolic values
- Extremely high diastolic values
To improve analytical accuracy:
- 1,023 patient records containing implausible blood pressure measurements were excluded from the Blood Pressure Relationship visual only.
- All remaining dashboard visualisations continue to use the complete dataset of 70,000 patients.
This demonstrates a practical approach to data validation while maintaining transparency.
- Data Cleaning
- Exploratory Data Analysis (EDA)
- Healthcare Analytics
- Calculated Fields
- Semantic Data Modelling
- Dashboard Design
- Interactive Filters
- KPI Development
- Histogram Visualisation
- 100% Stacked Bar Charts
- Scatter Plot Analysis
- Reference Lines
- Data Validation
- Business Storytelling
- Tableau Best Practices
Potential enhancements include:
- Patient risk scoring model
- Interactive BMI calculator
- Gender comparison dashboard
- Blood pressure category segmentation
- Geographic healthcare analysis
- Predictive machine learning integration using Python
This project is intended for educational and portfolio purposes.