Health Care in 3 Dimensions’ “Cube” is a solution for redirecting health care to changing behaviors. The interest in Artificial Intelligence as applied to health care is growing and expresses the concern if not amazement that technology has had little impact on major national policy goals of access and cost containment. In the JAMA article by Ezekiel Emmanuel June 18, 2019 “Artificial Intelligence in Health Care: Will the Value Match the Hype?” Dr. Emmanuel writes:

“A narrow focus on data and analytics will distract the health system from what is needed to achieve health care transformation: meaningful behavior change.”

The issue is not how to standardize behavior as one would expect to do with an analytic approach, but to protect behaviors that are beneficial and meaningful. The patient and physician N of 1 space respects the complexity of the patient that is not captured by population means. Uniqueness is the order of the moment and no patient was ever nor will be an average. Any good clinician will react to this uniqueness in a unique way, one patient at a time, with an eye to evidence based population medicine as conditional support when relevant.

Changing behaviors demands an approach that is customized. For unhealthy patient behaviors such as addictions, overeating, and sedentary lifestyles, there is not a metric that clarifies physician influence that is nuanced, persistent and effective for the given circumstance of the patient. For physician behaviors, interestingly defined as ordering tests, procedures, and pharmaceuticals, there is no metric that captures the numerous reasons physicians order these things. In fact it may be more useful to define these services as effects of behaviors, not the causes. If changing behaviors requires modifying effects, there will be an impasse with nothing happening. For example if overeating is an addictive drive it can be considered a mechanism leading to sleep apnea, then the solution is to manage the addiction. For the physician ordering a sleep study, which is an effect not a mechanism, the opportunity for cost saving is not to limit ordering sleep studies, but to leave the physician free to target the patient addiction and observe what many effects this would have. Thus we have 2 behaviors to note, the patient's with addictive overeating, and the physicians, with or without a propensity to dive deeper than simply recording a transaction.

To redirect health care as Dr. Emmanuel suggests, there needs to be a “Protected Space” for behaviors to play out without constraints by population level metrics or reimbursement requirements. If the behaviors are to be supported or changed, the measurement and evaluation activities must occur outside this Protected Space. Cost is a multidimensional effect of many behaviors, and the measurement when it focuses on the effects, can be translated in a meaningful way back to the protected N of 1 space where the mechanisms driving patient level costs can be addressed.

The HCn3D Cube is a vehicle for providing this Protected Space. Its structure with inputs from patient history, best practices, peer choices, patient event triggers, and a holistic provider mechanism puts a boundary around the moment of a physician decision or service. The Cube contains this moment which is called the Inflection Point in HCn3D. It is the demarcation between the past and the future. Policy goals can be placed in the future, and need not interfere in the moment. Outside the Protected Space all relevant features of the patient centered health care system are situated, incorporating the 2nd Dimension of provider networks, and the 3rd dimension of Future Value Outcomes. The Protected Space of the 1st dimension contains the data that affords AI, and all quantitative tools, the opportunity to identify and classify relevant metrics from the Patient Journey. This analysis occurring in the Protected Space provides the mechanism to influence behaviors without the risk of policy goals, such as cost containment, interfering with the moment.