After World War II, Taiichi Ohno helped develop the Toyota Production System, updating Kiichiro Toyoda’s ideas of Just in Time production to help Toyota close the performance gap between the company and top American manufacturers. Ohno defined seven types of waste:
1. Defective Production
2. Overproduction
3. Delay (i.e., waiting time in a queue)
4. Transportation
5. Inventory
6. Unnecessary movement or motion
7. Over processing (i.e., undertaking non-value added activity)
Mark Graban added an eighth waste, Human Potential.
I would frame Ohno’s model of waste in American health care in three distinct domains: addressing symptoms, managing chronic conditions and providing preventive care. This post will focus on using technology to minimize waste when managing a member’s symptoms and managing chronic conditions.
Using technology to direct members to the most appropriate site of service for initial evaluation
Today, members may be unsure of where to get evaluated when developing new symptoms. If a trusted health professional is not immediately available, a member may use whatever resources are readily available to make a decision about a formal evaluation. For many members, that means going to the emergency department. Patients with non-urgent conditions going to the emergency department for their care involves at least four types of waste (delay, transportation, unnecessary movement or motion and over processing). Weinick, Burns and Mehrotra found over 74.8% of emergency department visits for these conditions do not require emergency department care: upper respiratory infections, urinary tract and vaginal infections, conjunctivitis and other minor conditions (bug bites, rashes, contact dermatitis).
In 2014, Uscher-Pines and her colleagues reviewed 26 articles describing factors associated with non-urgent emergency department use. They found an average 37% (range 8-62%) of all emergency department visits were non-urgent. The researchers went on to develop a conceptual model of non-urgent emergency department use:
Uscher-Pines et al. suggested the causal pathway factors (red boxes) are stronger drivers of an individual’s decision to go to the emergency department for care than the associated factors (blue boxes). Any efforts to direct members with non-urgent conditions to other sites of care besides the emergency department should address convenience, cost, access/availability and provide knowledge and tools to help alter existing beliefs about alternatives.
In 2018, members have multiple ways to get information to help make a decision about engaging with the healthcare system. In addition to calling friends and family, members can access web browsers on mobile devices or traditional computers as well as query smart speakers. Those healthcare entities that provide tailored resources to address at least some of the causal pathway factors will be more effective at directing members to other sites of care for a diagnostic evaluation. Uscher-Pines et al. outlined five options when considering care (take no action, self-medicate, go to primary care provider, go to emergency department, go to other). With currently available technology, we should be able to tailor our triage efforts to provide even more options:
Triage systems that help members determine what conditions are emergent, what conditions are urgent and what conditions are routine should empower members to determine a more appropriate site of service for their initial evaluation. A more robust triage process could simultaneously reduce member waiting time, improve member satisfaction and decrease healthcare costs. An additional benefit of a triage system driving members to lower acuity care settings is that members will be subject to fewer tests or evaluations leading to subsequent healthcare utilization unrelated to the patient’s original symptoms.
Directing members to higher-value sites of service for surveillance or monitoring after a diagnosis has been made
Once a patient has a definitive diagnosis requiring follow-up (chronic condition), a similar array of service options could be used to help the member monitor the chronic condition. Patients with hypertension, diabetes, depression or other chronic conditions could be assessed through web surveys, chat bots, or other remote systems to keep the patient outside of a provider’s office.
As our monitoring technologies improve (vital signs and other physiologic parameters in the home, testing through urine or finger sticks), fewer chronic conditions should require face-to-face evaluations with healthcare providers.
This monitoring approach could rationally allocate limited face-to-face provider resources for monitoring behaviors that could be delegated to other members of the healthcare team. Patients deemed unstable or requiring complex care adjustments could be referred to a provider for their in-depth knowledge of the condition, treatment options and the implications of choosing each option.
Conclusion
Payors can use new paradigms and technologies to help engage their members in determining how they initiate care episodes as well as driving long-term management outside traditional care settings. Francois de Brantes of the Health Care Incentives Improvement Institute suggests health care spending can be divided into three zones: the retail zone (~15% of the total spend) including a large percentage of the population with relatively low costs; the manageable zone (~70% of total spend) with fewer patients but higher costs including chronic illness, acute conditions and procedures; and the insurance zone (~15% of total spend) including relatively rare instances of very high spending. Both interventions described above focus on targets to reduce waste within the manageable zone, the zone de Brantes suggests includes the largest savings opportunities.