How Pharma Companies Used AI to Map the Patient Journey and Identify Unmet Needs Across Key Markets

Introduction

Pharmaceutical companies are increasingly using artificial intelligence (AI) to understand how patients move through the healthcare system—from the first appearance of symptoms and diagnosis to treatment initiation, adherence, switching, and long-term disease management.

Traditional patient journey research often relies on surveys, interviews, claims data, medical records, and market research conducted independently. While these sources remain valuable, they can provide only partial views of a complex journey. AI can help connect signals across multiple datasets, identify patterns that may be difficult to detect manually, and segment patients according to behaviors, barriers, treatment experiences, and unmet needs.

For pharmaceutical companies operating across multiple markets, this capability is particularly relevant. Patient journeys can differ significantly because of differences in healthcare infrastructure, reimbursement, physician behavior, disease awareness, treatment availability, and socioeconomic factors.

This case study examines how AI-enabled patient journey mapping can help pharma companies identify unmet needs across key markets and translate those insights into opportunities for medical affairs, patient support, commercial strategy, and clinical development.

---

The Challenge: Patient Journeys Are Fragmented

A patient's healthcare journey rarely follows a linear path.

A typical journey may involve:

Symptoms → Self-management → Primary care → Referral → Diagnosis → Treatment selection → Treatment initiation → Monitoring → Adherence → Treatment modification → Long-term management

At each stage, patients can encounter barriers.

For example, a patient may experience symptoms for several months before seeking medical attention. After consulting a physician, the patient may undergo multiple diagnostic tests before receiving a confirmed diagnosis. Even after diagnosis, treatment initiation can be delayed because of affordability, reimbursement restrictions, physician preferences, or concerns about adverse events.

Pharma companies therefore need to understand not only what treatment patients receive, but also why they enter, leave, or progress through different stages of care.

The challenge becomes more complex when companies operate across markets.

A treatment barrier observed in the United States may not exist in Germany, Japan, India, or Brazil. Similarly, a diagnosis gap in one market may be driven by specialist availability, while another may be influenced by disease awareness or diagnostic infrastructure.

AI provides an opportunity to analyze these differences systematically.

---

AI-Enabled Patient Journey Mapping

Pharma companies can use AI to integrate and analyze multiple sources of patient and healthcare data.

These may include:

• Electronic health records
• Claims and reimbursement data
• Patient registries
• Clinical trial data
• Patient surveys
• Physician interviews
• Social listening data
• Patient support program data
• Treatment and prescription data
• Published literature
• Digital health interactions
• Disease-community discussions

Instead of analyzing each source independently, AI models can help identify relationships between different data points.

For example, an AI-driven analysis might identify that patients diagnosed with a particular chronic disease are frequently experiencing a long gap between diagnosis and treatment initiation. Further analysis could reveal that the delay is concentrated among patients who require specialist referral or face reimbursement restrictions.

This creates a more detailed view of the patient journey.

---

Case Study Objective

A multinational pharmaceutical company wanted to understand patient journeys for a chronic disease across five priority markets: the United States, Germany, Japan, India, and Brazil.

The company wanted to answer five questions:

1. Where are the largest patient drop-offs occurring?
2. What factors contribute to delayed diagnosis?
3. Why do eligible patients fail to initiate recommended treatment?
4. What drives treatment switching and discontinuation?
5. Which unmet needs differ by market?

The objective was not simply to create a visualization of the patient journey. The company wanted to identify actionable unmet needs that could inform market-specific strategies.

---

Step 1: Building a Unified Patient Journey Framework

The first step was to create a standardized framework that could be applied across markets.

The company defined six major stages:

1. Symptom recognition

2. Initial healthcare consultation

3. Diagnosis

4. Treatment selection

5. Treatment initiation and persistence

6. Long-term disease management

AI was then used to classify large volumes of qualitative and quantitative information according to these stages.

Natural language processing (NLP) could identify recurring themes in physician interviews, patient surveys, medical literature, and digital conversations.

For example, frequently occurring terms relating to "waiting," "specialist," "test," "cost," or "side effects" could be grouped into broader themes such as diagnostic delay, access barriers, or treatment concerns.

This helped transform unstructured information into structured patient journey insights.

---

Step 3: Using AI to Detect Unmet Needs

AI helped identify recurring unmet needs from multiple data sources.

The analysis grouped unmet needs into several categories.

Diagnostic unmet needs

Patients experienced delays caused by symptom overlap, limited disease awareness, and referral barriers.

Treatment unmet needs

Some patients required better tolerability, more convenient administration, or improved treatment outcomes.

Access unmet needs

Cost, reimbursement requirements, treatment availability, and healthcare infrastructure affected access in certain markets.

Information unmet needs

Patients and physicians sometimes lacked sufficient information about disease progression, treatment options, or long-term management.

Adherence unmet needs

Treatment complexity, adverse-event concerns, and insufficient follow-up contributed to discontinuation.

The value of AI was its ability to detect patterns across thousands of data points rather than relying exclusively on a limited number of interviews.

---

Step 5: Creating Patient Segments

AI also enabled the company to move beyond demographic segmentation.

Instead of simply dividing patients by age or geography, the company could identify behavioral journey segments.

For example:

The Delayed Diagnoser

Patients who experience prolonged periods between symptom onset and confirmed diagnosis.

The Treatment Hesitant Patient

Patients who receive a diagnosis but delay treatment initiation because of concerns about safety, cost, or treatment burden.

The Early Discontinuer

Patients who begin treatment but discontinue within a relatively short period.

The Treatment Switcher

Patients who move between therapies because of efficacy, tolerability, convenience, or other treatment-related factors.

The Long-Term Manager

Patients who remain engaged with treatment but require ongoing monitoring and support.

These segments could help pharma companies develop more targeted interventions.

---

Business Impact

The AI-enabled approach provided the company with a more granular understanding of patient behavior.

Instead of asking:

"Why are patients not using the treatment?"

the company could ask:

"At which stage are patients dropping out, which patient groups are affected, and what factors are associated with the drop-off in each market?"

This shift from broad assumptions to evidence-based journey analysis enabled more targeted decision-making.

The approach also helped reduce the risk of applying findings from one market universally to another.

---

Conclusion

AI is changing how pharmaceutical companies understand patient journeys. By combining structured and unstructured data, AI-enabled analytics can reveal where patients experience delays, treatment barriers, adherence challenges, and unmet needs.

The greatest opportunity is not simply to create more sophisticated patient journey maps. It is to use those maps to understand why patients behave differently, where healthcare pathways break down, and how unmet needs vary across markets.

For global pharmaceutical organizations, this approach can support more localized strategies across medical affairs, market access, clinical development, patient support, and commercial planning.

As healthcare data becomes increasingly interconnected, AI-enabled patient journey mapping is likely to become an important component of evidence-based decision-making—helping pharma companies move from broad patient population assumptions toward more detailed, market-specific understanding of real-world patient experiences.