Central Medical has an incentive to negotiate higher capitation rates, which will enable the practice to realize a financial return by aggressively managing utilization. Health First is motivated to $(1)$ set lower capitation rates because of the aggressive market, where businesses are willing to change insurers for small differences in cost$,$ and $(2)$ shift the risk to Central Medical. As a strategy, Central Medical enters negotiations with Health First, basing its utilization rates and costs on historical performance and taking a strong position on setting higher capitation rates. The practice believes patients will stay loyal to the clinics because of the high quality of staff and strong patient orientation. Both organizations accept their adversarial relationship, which they consider to be inherent in the healthcare and health insurance industries.

Central Medical's innovation team moves forward with developing a strategy for carrying out its mission. It focuses on disease management of complex chronic illnesses, such as type $2$ diabetes, as well as accessing and using evidence$-$based clinical guidelines. The practice discovers from available literature that intensive interventions, such as life coaches, demonstrate a $20$ percent reduction in glycosylated hemoglobin $($HbA$1$c$)$ within $6$ months for some patients. However, the team cannot justify the cost of adding the staff needed for intensive health maintenance for its sizable type $2$ diabetes patient population. This particular population is inherently high risk, although some patients with the same diagnosis use resources at a much higher rate than others. If the team could systematically identify diabetes patients who are at highest risk, it could better focus its intensive maintenance strategy, improve health, and increase efficiency. The challenge is identifying the characteristics of the patients with the highest risk in this high$-$risk population. These parameters are not revealed in the existing evidence from systematic reviews. Senior clinicians suggest relying on the clinical judgments of individual clinicians; although logical$,$ this approach cannot accurately identify the highest$-$risk patients because there are complex interdependencies. The team decides to analyze the collective medical records of its physician panel, but the predictive models lack rigor because of low population samples.

During a brainstorming session, the team proposes a collaboration with Health First to use Big Data to identify the highest$-$risk patients on the basis of their rate of resource utilization. From analyzing Health First's population$-$based enrollment data, the team is able to demonstrate that $1$ to $2$ percent of all patients with diabetes account for up to $30$ percent of the total costs for this diagnostic group. Variables used in the predictive models include total annual prescriptions, unique $($disease$-$specific$)$ annual prescriptions, physician visits, hospital utilization $($including emergency ser$-$vices$),$ comorbidities, age, gender, occupation, family composition, benefit coverage, and treatment history.

The team concludes that, by deploying a predictive model and a focused strategy, the information system could identify patients from a high$-$risk population who have the highest risk as well as a range of targeted interventions, such as life coaches and other intensive treatments.

One team member says, "Precision medicine from a managed care firm.

Who would believe it$!"$ Is the decision to provide intensive therapy for this high$-$risk population evidence based? Is there enough data in Central Medical's EHR to help identify and manage high$-$risk patients? How does it relate to clinical judgment?