DIABETES PATIENTS ANALYSIS REPORT

(An analysis of diabetes in Pima Indian women aged 21 and above)

Introduction

Diabetes is a medical condition that affects how the body regulates blood sugar (glucose). The dataset, originating from the National Institute of Diabetes and Digestive and Kidney Diseases, is designed to predict whether a patient has diabetes based on diagnostic measurements. It includes various medical predictor variables and one target dependent variable, “Outcome.” The dataset focuses on females at least 21 years old and of Pima Indian heritage.

About the Data

The dataset contains 768 rows and 9 columns. Its origin is from the National Institute of Diabetes and Digestive and Kidney Diseases. The main goal of this dataset is to predict whether a patient has diabetes, using various diagnostic measurements. The dataset exclusively includes female patients who are at least 21 years old and of Pima Indian heritage. There are multiple independent variables (medical predictor variables) and one dependent variable (Outcome) within the dataset.

Skills / Concepts Applied

The following skills and concepts were applied throughout the course of this project:

1. Defining KPIs
2. Power BI DAX Concepts: Calculated Measures
3. Data VisualizaBI
4. Power BI Dashboard building
5. Filters and Slicers
6. Insights & Actions

Data Transformation/Cleaning

The dataset was imported into Power BI’s Power Query for data validation and cleaning. ‘Column quality’ and ‘Column distribution’ checkboxes were selected to get summary information about each column for effective cleaning/Preprocessing. The processes are outlined below:

1. Created an ‘AgeGroup’ column using the following DAX code:

AgeGroup =
SWITCH(
TRUE(),
'diabetes'[Age] >= 21 && 'diabetes'[Age] <= 24, "21–24",
'diabetes'[Age] >= 25 && 'diabetes'[Age] <= 34, "25–34",
'diabetes'[Age] >= 35 && 'diabetes'[Age] <= 44, "35–44",
'diabetes'[Age] >= 45 && 'diabetes'[Age] <= 54, "45–54",
'diabetes'[Age] >= 55 && 'diabetes'[Age] <= 64, "55–64",
'diabetes'[Age] >= 65 && 'diabetes'[Age] <= 74, "65–74",
'diabetes'[Age] >= 75 && 'diabetes'[Age] <= 81, "75–81",
"Unknown"
)

2. Created Total Patients measure :

Total Patients = COUNT(diabetes[Age])

3. Created Diabetes Patients measure:

Diabetes Patients = CALCULATE('_measure'[Total Patients], 'diabetes'[Outcome] = 1)

4. Created Non-Diabetes Patients measure:

NonDiabetes Patients = CALCULATE('_measure'[Total Patients], 'diabetes'[Outcome] = 0)

5. Created Average Blood Pressure measure:

Average BloodPressure = AVERAGE(diabetes[BloodPressure])

6. Created Average Age measure :

Average Age = AVERAGE(diabetes[Age])

Insights and Visualization

The dataset reveals that diabetes patient distribution exhibits variations across a range of factors, encompassing age groups, glucose and insulin levels, BMI, and more. These findings suggest that diabetes is not confined to a particular demographic but affects individuals across diverse categories. The dataset comprises records for 768 patients, with 268 of them having a diabetes outcome, while the remaining 500 do not. On average, the patient’s age hovers around 33.24 years, and their mean blood pressure is approximately 69.11.

Additional insights extracted from the data are as follows:

1. Age Group Analysis: The age group 25–34 has the highest number of diabetes patients at 95, with a significant increase compared to the 65–74 age group, which has the lowest number at 4. This suggests that diabetes is more prevalent among individuals in their mid-20s to mid-30s.
2. Glucose Levels: Glucose levels are significantly higher in the 25–34 age group compared to the 75–81 age group.
3. Insulin Levels: The 25–34 age group has the highest sum of insulin at 16,915, indicating higher insulin requirements in this age group.
4. BMI and Diabetes: A BMI of 28 has the highest number of diabetes patients at 121. This suggests a correlation between higher BMI and the risk of diabetes.
5. Blood Pressure and Diabetes: A blood pressure of 92 has the highest number of diabetes patients at 70.
6. Pregnancies and Diabetes: The age group 25–34 with 86 pregnancies had the highest number of diabetes patients. This indicates the need for diabetes monitoring and education during pregnancy.

Recommendations

1. Public Health Awareness: Create public health campaigns to raise awareness about diabetes risk across all age groups. Emphasize the importance of regular check-ups and early detection.
2. Youth Education: Implement educational programs targeting individuals in the 25–34 age group, emphasizing the importance of a balanced diet, exercise, and regular monitoring of glucose and insulin levels.
3. Obesity Management: Develop weight management programs for individuals with a BMI of 28 or higher. Encourage healthy eating habits and physical activity to reduce the risk of diabetes.
4. Blood Pressure Monitoring: Promote regular blood pressure checks, especially for those with a blood pressure of 92 or higher, to reduce the risk of diabetes and its complications.
5. Pregnancy Diabetes Screening: Improve diabetes screening and education during pregnancy, focusing on the 25–34 age group.
6. Regular Health Check-ups: Encourage routine health check-ups for early diabetes detection and management, regardless of age or other factors.