Introduction

The battle to derive value from twenty-first century healthcare leviathans is just beginning. As the world and its economies become increasingly globalized, the struggle to balance rising costs with access to quality healthcare is not just a problem for highly advanced healthcare systems. It is a problem wherever resources are constrained, and process and bureaucracy trump relationships between human beings. Whether it is the connection between a teacher and student or a doctor and patient, a fundamental change is needed to place the person at the center of every decision regarding the best paths to their improved education or health. At the core of this transition is the need to measure the functional wellness of the patient and provide a tailored solution that meets their needs and objectives.

Today barriers exist within the healthcare system that limit the measurement and use of clinical outcomes and patient satisfaction rates that drive increased value. For example, one limitation is that human subject matter experts are often needed to review and assess medical data to determine whether a positive or negative outcome was achieved. Human experts are very good at identifying patterns based on selected indicators across administrative data, discharge summaries, or mortality/morbidity reports. They are also very good at determining the activities that added value over the entire course of treatment and recognizing those activities that did not bring value or caused delays in care. Additionally, the patients themselves know best whether their healthcare needs are being addressed. However, the incorporation of the human element in the value equation is time-consuming, labor and resource intensive, and may not produce actionable recommendations for a clinic or medical team for months or years.

A second limitation to ascertaining value in cycles of care for the individual patient is the growing volume and type of medical data that is available per person, per condition. One study estimates that by 2020 over 5,000 GB of data will exist for every person (Mearian, 2012). This includes medical reports, images, Electronic Health Record (EHR) data, and personalized sensors that produce real-time streams of everything from heart rate to pain thresholds. Key clinical data in these records is typically recorded in unstructured form as free text and images; most structured clinical information is poorly organized. Time-consuming interpretation and analysis is required to convert these records into structured clinical data (Rao, 2003). Today, the use of even the most common medical data fields to routinely assess value in the delivery of care is only in the research phase. In only a few more years, a human expert alone will not be able to effectively analyze, consolidate, and identify crucial patterns in all of the available data on a patient.

The traditional procedures that exist to share and use clinical outcome data across the healthcare enterprise are another obstacle. At the macro level, many institutions promote best practices and communicate evolving standards and clinical pathways in the treatment of disease. However, in today’s complex medical microenvironment there is a constant movement either closer to or farther from the desired quality outcome(s). A continual feedback mechanism across the enterprise to effectively gather, share, and measure value — in improved quality and reduced costs — is severely lacking. This leads to a stagnant and uninspired organization that does not acquire or transfer knowledge.

Blood Disease Management in Iraqi Kurdistan

Agile HMS is fighting the same fundamental battle as organizations in advanced healthcare systems: the need to demonstrate value in their Population Health Management services. Agile primarily does business in India, Iraqi Kurdistan and Kenya, and uses the Orsalus Exchange (OEX) as a patient and disease management platform. The following is a real world example from this partnership. Agile has thousands of patients suffering from thalassemia in Kurdistan that are managed in three clinics located in the geographically dispersed cities of Erbil, Sulymania and Dohuk. The primary treatment is regular blood transfusion and oral medication to combat iron overload. A bone marrow transplant (BMT) is the only solution that permanently cures the patient.

When Agile started work in Kurdistan, each city clinic worked in isolation. There was no systematic data collection or analysis. There was no insight into how a patient’s history and current condition contributed to their overall health outcomes. There was little understanding of what factors increased or decreased the costs of a BMT. After surgery, there was little follow-up data on whether patients survived, thrived or were defeated by their illnesses. After applying a systematic process for collecting data throughout the individualized cycle of care, Agile – using Orsalus software – has now been able to identify activities that add value and, through automated analysis techniques, improved the processes used for delivering and deciding on treatment options. Specifically, the approach has been able to impact the following patient outcomes:

1. Increased access to bone marrow transplants by over 1,000% (10 annual treatments prior to 2013, compared to more than 100 annual treatments in 2015) by improving the collection and use of patient data in treatment plans.

2. Reduced costs of available treatments by more than 30% through capacity management, volume discounts based on the true cost of care, and process improvement.

3. Categorized patients into high / low risk and established basic forecasts on the full cost of care for these patients through automated analysis.

4. Identified 2 non-related donors within the donor-database created to support the program, for patients lacking a sibling with matching HLA-typing.

5. Extended data collection and analysis processes to the management of malignant blood disease, non-malignant blood disease, chronic kidney disease, and liver disease.

This on-going effort includes 1) the collection and standardization of data elements (over 150 individual data points per patient) to assess positive and negative impact on patient outcomes and costs, including patient survey data; and 2) the development of models such as binary classification to sort patients into high / low risk categories and identify value added activities utilizing cloud-based processing tools. Orsalus believes strongly that the approach taken for thalassemia is directly applicable to all healthcare systems. The analytical approaches applied to the thalassemia dataset can serve as the starting point for other disease dataset. However, to truly impact how cycles of care for all conditions are delivered, measured, and reimbursed in the near term it is necessary to advance the algorithms being used today.