We’ve come a long way from “take two and call me in the morning.”

Advancements in genomics, biomarkers, and artificial intelligence (AI) have enabled precision medicine, allowing clinicians to prescribe treatment plans based on an individual’s specific biology. Now, the rapid proliferation of data is pushing personalized medicine even further.

Abundant, rich data sources, such as multiomics and wearables, are giving clinicians deeper insights to support first-line treatment decisions. Integrated biological, behavioral, and clinical history is helping physicians identify the right therapies with far greater precision than genetic testing alone. And clinical data is being collected almost constantly, allowing for around-the-clock micro-adjustments.

Instead of using data only to choose a therapy, some physician teams and patients are using data to optimize treatments in real time.

Take chronic kidney disease as an example. A traditional treatment plan might include a fixed dose prescription for an ACE inhibitor, with lab orders to return in three to six months. When choosing a therapy, a clinician may account for typical disease trajectory, statistical risk factors, or the patient’s biomarkers; however, the treatment plan is still anchored in trial and error. And meaningful disease progression can occur during the “wait and see” period.

Now, longitudinal data about disease progression, risk factors, and biomarkers can be layered on top of real-time data. If a patient’s weight or blood pressure shifts, even subtly, a digital health platform can trigger an immediate intervention.

This dynamic optimization is the next frontier for personalized medicine. And in some therapeutic areas, it’s a potential risk for pharma.

The data challenge for pharma  

In the past, pharma companies have tried to insert themselves into care management with little success. Apps intended to influence patient education and treatment adherence have seen 3%–5% retention rates after 30 days and delivered negative return on investment. These efforts failed because pharma companies attempted to own patient services rather than the relationship with data or infrastructure. 

As care becomes continuous and data-driven, value is shifting toward companies that influence or integrate with data, algorithms, and patient interfaces—not just the therapies. Today, the most valuable data sits close to the patient (e.g., in electronic health records [EHRs], provider platforms, and wearables), not with pharmaceutical companies.

This is a critical disadvantage. Without stronger access to these assets, pharma could lose critical influence over important treatment decisions. Forward-thinking leaders are finding ways to innovate beyond the molecule, developing precision treatment algorithms and codifying them into protocols and decision tools. They’re also seeking ways to embed themselves in post-prescribing infrastructure to track adherence, drug switching, and outcomes.

With stronger access to data, pharma can help shape the patient journey long after a script is written. In some cases, the nature of the therapy itself creates the opportunity. Take Novartis’s Kymriah^®, a CAR-T cell immunotherapy, for example. The therapy can only be delivered through a highly controlled care model, requiring designated treatment centers, strict protocols, and intensive monitoring for severe side effects. Providers report adverse events and collect patient samples through defined protocols, creating a continuous data loop between the care setting and manufacturer. Rather than simply supplying the drug, Novartis is embedded in how treatment is delivered and managed after prescription.

In another example, Gilead Sciences engineered adherence into its Sunlenca^® therapy for patients with multidrug-resistant HIV. The therapy replaces daily pills with an injection administered every six months, turning adherence into a managed process rather than a daily patient decision.

Access to enhanced data capabilities also gives pharma companies a direct lever to address two of their biggest commercial vulnerabilities: competitive switching and nonadherence. Companies with multiple products in a disease area need data to manage patients across the portfolio and keep them in the franchise.   

Breaking the trial-and-error cycle

Data is likely to drive the most value for conditions defined by episodic or cyclical care, such as obesity, diabetes, mental health, fertility, and hypertension.

In these areas, patients often cycle through treatments, waiting weeks or months to see if a therapy works before switching to new one. That process is repeated until a patient finds the right fit or quits. On average, 65% of patients with chronic diseases such as type 2 diabetes, rheumatoid arthritis, asthma, and attention-deficit/hyperactivity disorder switch therapies or discontinue treatment during their journey.

Data has the power to break that cycle. For example:

• In oncology, AI enables faster biomarker detection and resistance monitoring.
• In immunology, algorithms can predict treatment responses, helping modulate doses to prevent flares.
• In obesity, data analytics can predict fallout and nonresponse risk, triggering tailored coaching or medication changes before a patient drops out.
• In diabetes care, patients already adjust treatment regimens and doses based on self-monitoring devices rather than static physician orders.

Dosing, sequencing, and drug combinations are becoming dynamic rather than fixed across a range of chronic diseases. With the right data and tools, disease management can become continuous and predictive rather than reactive.

Choosing a strategic position

The pace of this evolution is uncertain and will play out differently across therapeutic areas. Pharma leaders must decide what role they will play as data becomes as powerful as the drugs.

Right now, there are three available paths:

• Suppliers continue to influence treatment decisions through established channels but have limited say in how therapies are used and adjusted post-prescription.
• Solution developers embed evidence into clinical decision support tools and workflows. These companies provide services and data that shape treatment decisions, adherence, persistence, and outcomes.
• Ecosystem orchestrators leverage data, partnerships, and capabilities to shape the end-to-end patient journey.

The “right” model will depend on a firm’s specific portfolio, therapeutic area, and capabilities.

Notably, no single player can own this shift alone. Continuous precision care requires coordination across an ecosystem of both old and new players, encompassing development, treatment, and reimbursement. Pharma leaders need to stake a position of influence in the ecosystem early to prevent being “optimized out” of system- and algorithm-driven care decisions.

How to win in precision care

To win in precision care, pharma companies need to influence key parts of the care ecosystem. Leading companies are beginning this transformation now, building capabilities and partnerships that combine data and clinical intelligence to inform real-time, adaptive interventions. They are:

• Redesigning trials around adaptive and patient-level endpoints. Instead of waiting for the end of a study to analyze a fixed group, some companies are adjusting trials in real time based on how individuals respond. For example, Roche used wearable sensors and a smartphone app to collect daily gait analysis for multiple sclerosis (MS) patients, reducing reliance on the traditional six-month Expanded Disability Status Scale with continuous, real-world data.
  
  
• Structuring data for AI-driven decision environments, ensuring that both clinical and real-world evidence can be surfaced, interpreted, and prioritized by algorithms at the point of care.
  A major hurdle in AI-driven care is that data is often unstructured or multimodal. To solve this, companies like Tempus AI have built operating systems that integrate genomic, transcriptomic, imaging, and EHR data. This allows clinicians to receive guideline-based treatment recommendations that keep pace with medical advancements. Similarly, AstraZeneca is piloting technology to automate the flow of data between EHRs and clinical trial systems. Removing manual transcription supports stronger data availability, quality, and accuracy. It also means trial evidence can be interpreted faster and with less burden on staff.
  
  
• Equipping commercial teams to support treatment management beyond the point of prescription. Teams need the tools and skills to help clinicians and patients manage adherence, drug switching, and treatment optimization over time. In the future, the most sophisticated organizations may go further: using longitudinal patient and outcomes data to identify drop-off risk, trigger interventions, and maintain visibility into how therapies perform in the real world.

The window for action is closing

The authors would like to acknowledge and thank the broader team that contributed to developing this point of view, including Brittany Rodriguez, Kristin Moneyron, Kusha Korla, Vipresha Jain, Anirudh Mishra, and Anusha Prasad.