Learning Objectives

By the end of this lesson, you should be able to:

• Understand how healthcare systems are organized.
• Understand the flow of patients through the healthcare system.
• Identify the major sources of healthcare data.
• Recognize common healthcare stakeholders.
• Understand where a healthcare data analyst fits into the system.

---

Why Learn Healthcare Systems?

Many new healthcare analysts make the mistake of jumping directly into:

• SQL
• Dashboards
• Machine Learning
• Statistics

without understanding how healthcare actually works.

Imagine analyzing diamond sales without understanding:

• suppliers,
• retailers,
• deployment,
• sales,
• inventory.

You could compute numbers but would struggle to answer meaningful business questions.

Healthcare is the same.

Before analyzing healthcare data, you need to understand:

Who generates the data?

Why does the data exist?

How does the patient move through the system?

---

The Big Picture

Healthcare systems exist to:

1. Prevent disease
2. Diagnose disease
3. Treat disease
4. Monitor recovery

The patient is at the center of the system.

A simplified view:Patient ↓ Primary Care ↓ Specialist ↓ Hospital ↓ Recovery / Follow-up

Every interaction creates data.

---

Main Components of Healthcare

1. Primary Care

This is usually the first point of contact.

Examples:

• Family physicians
• General practitioners
• Nurse practitioners

Data Generated

• Blood pressure
• Weight
• Height
• Medical history
• Referrals

Example:

A patient visits a family doctor because of fatigue.

The physician records:

• age
• sex
• symptoms
• blood pressure

This becomes part of the Electronic Health Record (EHR).

---

2. Specialists

When a problem requires advanced expertise.

Examples:

• Cardiologists
• Oncologists
• Neurologists
• Endocrinologists

Data Generated

• Diagnoses
• Specialist notes
• Procedures
• Test results

Example:

The family physician suspects heart disease.

The patient is referred to a cardiologist.

New data are generated:

• ECG
• Echocardiogram
• Cardiac diagnosis

---

3. Hospitals

Hospitals handle more complex care.

Examples:

• Surgery
• Emergency visits
• Intensive care

Data Generated

• Admissions
• Procedures
• Medications
• Lab tests
• Discharge summaries

Example:

Patient experiences chest pain.

Hospital records:

• Admission time
• Blood tests
• Imaging
• Medications

---

4. Pharmacies

Pharmacies dispense medications.

Data Generated

• Prescriptions
• Refill dates
• Drug utilization

Example:

Patient receives:

• Metformin
• Insulin

Medication records become valuable for analytics.

---

5. Public Health

Focuses on populations rather than individuals.

Examples:

• Vaccination programs
• Disease surveillance
• Outbreak monitoring

Data Generated

• Infection rates
• Vaccination rates
• Mortality rates

---

Patient Journey Example

Consider a patient named Sarah.

Stage 1

Sarah feels tired.

She visits her family physician.

Data collected:

Variable

Value

Age

52

Weight

78 kg

Blood Pressure

145/95

Symptoms

Fatigue

---

Stage 2

Doctor orders blood tests.

Results:

Test

Result

Glucose

High

HbA1c

High

Possible diabetes.

---

Stage 3

Referral to endocrinologist.

Additional data:

• Diagnosis
• Treatment plan

---

Stage 4

Prescription issued.

Medication data generated.

---

Stage 5

Follow-up visits.

Additional outcomes recorded.

---

Healthcare Stakeholders

As analysts, we rarely work directly for patients.

We usually work for stakeholders.

Patients

Interested in:

• Better outcomes
• Lower costs
• Better experiences

---

Physicians

Interested in:

• Better treatment decisions
• Faster diagnosis

---

Hospitals

Interested in:

• Capacity
• Resource allocation
• Quality metrics

---

Governments

Interested in:

• Population health
• Cost control
• Policy effectiveness

---

Insurance Companies

Interested in:

• Risk
• Claims
• Fraud detection

---

Types of Healthcare Analytics

Healthcare analytics generally falls into four categories.

---

Descriptive Analytics

“What happened?”

Example:

• Number of admissions
• Average length of stay

---

Diagnostic Analytics

“Why did it happen?”

Example:

Why did emergency visits increase?

---

Predictive Analytics

“What will happen?”

Example:

Will a patient be readmitted?

---

Prescriptive Analytics

“What should we do?”

Example:

How many nurses should be scheduled next week?

---

Where Data Analysts Fit In

Healthcare analysts support decision-making.

Examples:

Hospital Analyst

Questions:

• Which departments are overcrowded?
• What is average length of stay?

---

Population Health Analyst

Questions:

• Which communities have high diabetes rates?
• Which interventions work?

---

Clinical Analyst

Questions:

• Which treatment produces better outcomes?

---

Operations Analyst

Questions:

• How many beds are needed?
• How many nurses should be scheduled?

This area overlaps heavily with Operations Research.

---

Common Healthcare Metrics

Mortality Rate

Percentage of patients who die.

\text{Mortality Rate}=\frac{\text{Deaths}}{\text{Population}}

---

Readmission Rate

Percentage of patients returning after discharge.

\text{Readmission Rate}=\frac{\text{Readmitted Patients}}{\text{Discharged Patients}}

---

Length of Stay

Number of days spent in hospital.

\text{LOS}=\text{Discharge Date}-\text{Admission Date}

---

Bed Occupancy

Percentage of beds currently occupied.

\text{Occupancy}=\frac{\text{Occupied Beds}}{\text{Available Beds}}

---

Real Example: Hospital Capacity Problem

Suppose a hospital has:

• 500 beds

Current occupancy:

• 475 beds occupied

Occupancy:

$$\frac{475}{500}=0.95=95%$$

Questions management may ask:

• Will capacity be exceeded next month?
• Should additional staff be hired?
• Should elective surgeries be postponed?

A healthcare analyst helps answer these questions.

---

Key Takeaways

1. Healthcare systems consist of:
   • Primary care
   • Specialists
   • Hospitals
   • Pharmacies
   • Public health
2. Every patient interaction generates data.
3. Healthcare analytics answers:
   • What happened?
   • Why did it happen?
   • What will happen?
   • What should we do?
4. Healthcare analysts support:
   • Clinical decisions
   • Operations
   • Resource planning
   • Population health
5. Understanding the patient journey is essential before learning healthcare databases, SQL, statistics, or machine learning.

---

Exercise

Draw the healthcare journey for a patient with diabetes:

1. Family doctor visit
2. Blood tests
3. Endocrinologist referral
4. Medication
5. Follow-up visits

For each step, write down:

• Who generated the data?
• What data was generated?
• Why was the data collected?

This exercise will prepare you for Lesson 2: Medical Terminology for Data Analysts, where we’ll learn how diagnoses, procedures, and diseases are represented in healthcare data.