How automated chronic disease management can help scale care

A recent study from the Brigham and Women's Hospital demonstrates exciting possibilities.
By Dr. Daniel Yazdi

About the Author: Dr. Daniel Yazdi is an Internal Medicine Resident at Brigham and Women’s Hospital/Harvard Medical School interested in the intersection of digital health and clinical medicine. He can be followed on Twitter @DanielYazdi.

Medicine is reaching a problem of epic proportions: without any changes to how primary care is delivered, in 2030 there will be a projected deficit of 14,800 to 49,300 (25th to 75th percentile) primary care physicians (PCPs). Chronic diseases (such as heart failure, hypertension, diabetes) are managed by PCPs and take up a majority of healthcare spending that accounts for 17% of the United States GDP. The problem is a supply and demand mismatch: too many patients, too few physicians. Historically, other industries have overcome this problem by utilizing structured “algorithms” to make product development more scalable, efficient, and with minimized variation in quality. Can medicine undergo its own “Industrial Revolution” to make patient care more algorithmic, scalable, cost-effective, while optimizing patient outcomes and reducing its variability?

Physicians at the Brigham and Women’s Hospital have been investigating this proposition. In a recent publication in the Clinical Cardiology Journal, cardiologist Dr. Benjamin Scirica et al. discuss a novel remote care-delivery paradigm for hypertension management led by non-medically licensed patient navigators, termed Brigham Protocol-based Hypertension Optimization Program (BP-HOP). As described in this article, their paradigm has the potential to shift our approach to chronic disease management.

Computerized Algorithm-Driven Care

At its core, the program is founded on a clinician-driven algorithm for blood pressure management based on national treatment guidelines. The algorithm contains a step-by-step approach for medication initiation, patient follow-up and medication titration. Patients are provided with a Bluetooth blood pressure cuff to measure their blood pressure at home. Under the general supervision of licensed pharmacists, the non-licensed clinical navigators are able to remotely monitor and engage in telephone communications with their patients to improve their hypertension regimen. Of the 130 patients enrolled in the study, 81% achieved adequate blood pressure control in a remarkable average of seven weeks. This was achieved without an increase in pill burden (they used combination drug pills) and with continued control during the one-year follow up after patients “graduated” from the program.

Template for Other Chronic Diseases

Management of many chronic conditions — including diabetes, hyperlipidemia (high cholesterol) and heart failure, to name only a few — are driven by a standardized set of medical society guidelines. This allows for the similar construction of algorithms that can codify the management of these conditions. The positive ramifications are manyfold, including efficient and rapid management of the condition, improved patient experience and a scalable solution that can reduce the burden on primary care physicians and the US economy.

Efficient and Rapid Disease Management

Given the ability for the navigators to remotely monitor patients via Bluetooth-enabled blood pressure cuffs and telephone calls, the optimal blood pressure medication can be identified in the shortest possible time. This bypasses the inefficient traditional clinical model where patients schedule follow-up appointments to have their blood pressure measured and medications adjusted. Navigators have the sole responsibility of optimizing the patient’s blood pressure, something that can be delayed or overlooked when only managed by a PCPs given their other demanding responsibilities.

Improved Patient Experience

Managing chronic diseases from home has many advantages for patients. Patients do not need to take time off from work, find transportation to the clinic, spend money for parking or any of the other burdens associated with attending a doctor appointment. The navigators in this study continued to follow the same patients, allowing for the development of a patient-navigator rapport and creating a sense of accountability for the patients. Furthermore, phone calls can be scheduled at times most convenient for the patient.

Reduction in Physician Demands

Management of chronic diseases in this fashion can alleviate a significant burden from the already demanding jobs of PCPs. Most PCPs only have 15 minutes per patient visit, making it nearly impossible to tackle multiple chronic conditions, especially if new issues are being discussed during the visit. To combat this, PCPs often schedule frequent follow-up appointments which further stresses the medical system. With an algorithm-driven navigator program, PCPs can address fewer and more acute concerns during clinic visits. With fewer problems to tackle each visit, PCPs can potentially increase their panel size, improving access to care.

Scalable Solution

This model for chronic disease management is scalable since the underlying medical knowledge is codified in a computerized algorithm. Non-licensed patient navigators will take care of a majority of the patient interactions. This results in a reduced dependency on the limited supply of physicians, nurse practitioners and pharmacists to manage chronic diseases, thereby making care more accessible. Furthermore, the infrastructure is inherently scalable, using telephone communications, electronic health records and wireless home sensors.

Economic Implications

With the passage of the Patient Protection and Affordable Care Act of 2010, there is an increased shift towards bundled payment plans and accountable care organizations that aim to provide high-quality care at low costs per patient per year or per episode of care for a given condition. A computerized, algorithm-driven navigator program aligns with these incentives given its scalable nature, time-sensitive obtainment of the desired health outcomes, and therefore increased likelihood of mitigating co-morbidities and hospitalizations associated with poorly managed disease.

Similar Solutions and Future Challenges

There are many chronic disease management platforms including Twine Health (acquired by Fitbit) and WellDoc, but what makes this system novel is its detachment from direct physician involvement. Programs like Twine require physician approval prior to any medication changes suggested by the health coaches. A computerized algorithm removes the already taxed physician from direct management decisions. However, no algorithm is perfect and medically licensed practitioners need to be available when patients “fall off” the decision tree. Such a program will also need to be seamlessly integrated within the electronic health records so that the entire care team can be aware of the changes. The remote monitors must require minimal installation to maximize patient utilization and adoption. If these challenges can be overcome, algorithmic chronic disease management has the potential to revolutionize medicine during this most critical time of need.