Prescription Under Dual Control: How Artificial Intelligence Helps Physicians Avoid Errors

In outpatient clinics and hospitals across the Moscow Region, a new digital assistant has begun supporting physicians. The AI-based module activates at the moment a prescription is drafted and rechecks medication orders before the document is finalized and handed to the patient.

The system is embedded directly into the region’s medical digital framework. As soon as a physician enters a drug into the patient’s electronic record, the algorithm automatically analyzes it against approximately 15 key parameters. It evaluates compatibility with other prescribed medications, considers the patient’s age and sex, flags potential overdosing, reviews treatment duration and identifies other clinical risks. If a potential issue is detected, the module alerts the physician before the prescription is formally issued.

The final decision remains with the physician. The algorithm does not make independent clinical determinations; it functions strictly as an intelligent assistant. In high-volume settings with heavy patient flow, the AI helps clinicians avoid overlooking critical details.

Clinical and Patient Benefits

The new tool is used not only during in-person appointments but also in telemedicine consultations. For example, it supports cases where a patient needs to renew or adjust a subsidized prescription. This is particularly relevant for individuals with chronic conditions who require ongoing pharmacological support.

Deploying such systems marks a step toward the medicine of the future, where technology serves not as a replacement but as a complement to a physician’s knowledge and experience. It is important that the module functions strictly as an assistant; the decision to prescribe a medication remains with the specialist

Maxim Zabelin

Deputy Chairman of the Government of the Moscow Region – Minister of Health of the Moscow Region

For patients, the primary benefit is safety. Most individuals cannot independently assess drug interactions or verify whether a dosage is appropriate. When a prescription undergoes an additional digital review, the likelihood of accidental error decreases.

The system also strengthens trust. Patients understand that their treatment is subject to dual oversight – by the physician and by a data-driven system capable of analyzing large datasets and accounting for variables that may be missed due to fatigue or time pressure.

Raising the Quality of Care

In the Moscow Region, 67 AI-driven projects have already been implemented, with another 15 scheduled for launch this year. The clinical decision support module for prescription review has become part of this broader digital health infrastructure.

The project did not emerge in isolation. In recent years, regions across Russia have systematically introduced digital tools into healthcare delivery. Diagnostic assistants are developing in parallel, including the system AIDA, which helps reduce the risk of diagnostic errors.

For Russia, deploying such algorithms represents a shift toward more systematic quality management in healthcare. The approach enables analysis of anonymized data on prescribing practices, identification of recurring risk patterns, refinement of clinical guidance and improved therapy safety.

Scaling Within Russia and Beyond

Global research indicates that clinical decision support systems can recommend treatment pathways with a high level of adherence to professional guidelines. Russia is aligning with this trend by developing domestic solutions tailored to its national electronic health record infrastructure and care standards.

Within the country, the next logical step is scaling similar modules to additional regions and integrating them into larger medical platforms. As data exchange standards become more unified, algorithms will operate with greater precision and speed.

The international market also presents opportunities. Many countries with comparable healthcare systems are seeking tools that reduce prescribing errors without requiring a fundamental overhaul of infrastructure. Export potential lies not only in the software itself but also in the implementation expertise gained through integration into real-world clinical workflows. Developing an algorithm is only part of the challenge; embedding it into everyday practice so that physicians use and trust it is equally critical. Russia has demonstrated strong capabilities in this area.