The Logistics of Population Health Under the IHI Triple Aim

July 17, 2015 / By L. Gordon Moore, MD

The Triple Aim is a construct developed to move past the trade-offs typical in health care improvement: improved quality at the expense of increased costs, decreased costs at the expense of quality or access, improved guideline adherence at the expense of patient experience of care, etc. The Triple Aim defines success as simultaneous improvement in population health and outcomes, patient experience of care, and cost trends.

Because of the immense pressures to reduce per capita medical expenditure in the near-term, Triple Aim strategies must also succeed in the near-term. The list of potential improvement strategies is far longer than the capacity to make near-term change, so it is logical and appropriate to consider those strategies most likely to achieve the goals at the top of a priority list.

This prioritization challenges deeply held beliefs regarding delivery system change. Preventive care – while essential – is not likely to achieve near-term cost savings. The outcomes of tight blood sugar control in diabetes (fewer heart attacks, amputations, strokes) show up nine years later. [Holman 2008] Closing gaps in evidence-based care through the use of clinical registries is necessary but insufficient. We must do more if the Triple Aim is our goal.

Improving population outcomes means more than just aggregating our medical record data on those patients we see in our practices: our responsibilities extend to the population attributed to the practice and their health system interactions beyond our four walls. It is important to know about patients who fail to come to our offices but show up in multiple emergency departments, who bounce from specialist to specialist, and who have chronic conditions but are receiving no care. Claims data provide a window into the people and activity beyond the practice.

A segmentation strategy focused on the entire population rather than a specific diagnosis provides a pool of opportunity more likely to achieve the Triple Aim. Interventions that simultaneously improve outcomes, experience of care and cost are more likely to achieve the Triple Aim than interventions addressing only one or two of the axes. Below are segmentation and intervention strategies that meet these criteria. This not a complete list of all potential interventions and is not meant to be exclusive of other interventions.

Eventually we may be able to support all the needs of all patient segments, but as we get started it makes sense to begin with very high risk individuals and expand as our capacity allows. A number of groups start with a patient engagement strategy. This begs the questions: who do we engage first?

Narrowing down to a patient segment – “population of focus”

Creating a “hot-spotter” initiative

The “hot-spotter” strategy identifies people with extraordinary risk (high illness burden + high ER/hospital admission use). [Gawande 2011] Care management teams can use this data to identify high risk patient groups and

A cadre of care management staff (a.k.a. health coach, care manager, case manager) reviews the chart with the PCP, reaches out to the individual, tailors interventions to support that person’s needs.

Outreach to loosely or un-engaged individuals

Being responsible for a population means reaching out to loosely or un-engaged people with complex chronic conditions. A care management team can run monthly reports from our dashboards identifying individuals with chronic conditions who have no PCP visit in 6 months or no physician visit at all in 12 months. Lists can also be run to identify people with worsening clinical conditions.

Patient activation

Claims and clinical data are only part of the more complex reality of our patients. “Non-compliance” is a set of discoverable opportunities for intervention. Some in this category have competing personal agendas and choose not to work on their condition at this time. They would benefit from stages-of-change or motivation interviewing types of interventions.

Individuals who are interested in working on their condition(s) but are unsuccessful benefit from interventions targeting self-management support/patient self-efficacy. [Hibbard 2013] Achieving better outcomes for this population segment requires embedding a systematic means for unmasking patient-reported risk factors. Practices can do this by using a health risk assessment and integrating its results with those from other reporting tools. [Wasson 2011]


This has been a very high level overview of some strategies that are likely to simultaneously improve patient health and outcomes, experience of care, and cost trends. These strategies can be added to the mix of programs in a health care delivery system and are especially important in settings where reduced cost trends through better care delivery is the goal.

L. Gordon Moore, MD, is senior medical director for populations and payment solutions at 3M Health Information Systems.

Gawande, Atul. “The Hot Spotters.” The New Yorker, January 24, 2011.

Hibbard, Judith H., and Jessica Greene. “What The Evidence Shows About Patient Activation: Better Health Outcomes And Care Experiences; Fewer Data On Costs.” Health Affairs 32, no. 2 (February 1, 2013): 207–214. doi:10.1377/hlthaff.2012.1061.

Holman, Rury R., Sanjoy K. Paul, M. Angelyn Bethel, David R. Matthews, and H. Andrew W. Neil. “10-Year Follow-up of Intensive Glucose Control in Type 2 Diabetes.” New England Journal of Medicine 359, no. 15 (October 9, 2008): 1577–1589. doi:10.1056/NEJMoa0806470.

Wasson, John H, Regina Benjamin, Deborah Johnson, L Gordon Moore, and Todd Mackenzie. “Patients Use the Internet to Enter the Medical Home.” The Journal of Ambulatory Care Management 34, no. 1 (March 2011): 38–46. doi:10.1097/JAC.0b013e3181fff641.