What you need to know about the forces reshaping our industry.


Emerging Idea

Precision Population Health

10 Minute Read


The idea

Precision population health (PPH) is a sophisticated and predictive form of risk stratification that borrows concepts from precision medicine and population health management. PPH models provide nuanced risk profiles of patients who may or may not routinely access health care services. Providers can then proactively connect at-risk patients to targeted, high-impact health or social care interventions before the patients escalate.

The promise

Precision population health will help organizations target finite resources across an increasingly more complex patient population in two ways. First, it will pinpoint cohorts of patients with risk factors that can be reduced. Second, precision population health will make it easier to identify the clinical or non-clinical solutions that yield the biggest impact on health outcomes and total cost of care.

Why now

Many patients are deferring care and adopting unhealthy behaviors amid the Covid-19 pandemic. Doing so exacerbates preexisting conditions and creates blind spots in providers’ knowledge of their populations’ health needs. Health care organizations are also facing unprecedented financial pressures, putting a premium on health care solutions that lower total cost of care.

Reality check

Precision population health requires substantial investment in data repositories, technology, staff, and infrastructure. Many providers may not have these assets in-house and may lack the funds to purchase them. Further, precision population health requires a high level of coordination across all downstream sites, social care, and local payers to ensure that patients are connected to the most impactful clinical and non-clinical interventions. This level of coordination is historically uncommon.


What is it?

Precision population health (PPH) is a sophisticated and predictive form of risk stratification that borrows concepts from precision medicine and population health management.

Precision medicine and PPH each identify patients’ specific health care needs that are otherwise difficult to uncover and provide a set of targeted interventions to address those specific needs. The main difference between precision medicine and PPH is scope. Using expansive clinical, psychosocial, and public and proprietary consumer data sets, PPH proactively identifies patient cohorts with similar risk profiles that may escalate to high-cost care. Through PPH, health systems can then proactively connect these cohorts to the specific clinical and non-clinical interventions that will yield both the greatest benefit to the cohort’s health outcomes and biggest cost reductions to the system. In contrast, precision medicine targets individual patients for personalized treatments using genomic medicine.

PPH also builds on existing population health management frameworks to target a narrower cross-section of patients. Traditional risk stratification frameworks categorize patients as low-, moderate-, and high-risk based on disease burden, utilization, and cost. PPH is more precise in calculating patient risk, as well as identifying and grouping patients into cohorts. Additionally, due to their expansive data sets and computing power, precision population health models can identify any patient in an organization’s network who is at risk for a clinical or non-clinical need, not just those with whom providers have frequent interaction.


Why now?

Precision population health is emerging as systems are becoming more open to value-based payment models to curb the rising costs of care. While we’ve yet to see the full extent of how Covid-19 will impact the shift to value-based care, precision population health accomplishes many of the same goals: improving health outcomes, reducing total cost of care, providing proactive care, reducing inappropriate demand for treatment, and uncovering latent or undiagnosed illnesses.  

IT capabilities are catching up to our understanding of patient risk

Precision population health has only recently become possible because of broader access to clinical and consumer data, as well as recent advancements in computing power.

Over the past decade, large data repositories (“big data”) have become far easier and cheaper to access. With this shift, population health management models have become more sophisticated as progressive, value-leaning organizations have invested in and incorporated more nontraditional data sets (e.g., credit score, purchasing habits, internet history and access) into their risk analyses. Layering on this data allows providers to better capture root causes of patient risk and more accurately inform resource allocation to solve for them.

Further, recent advancements in computing power have allowed these organizations to gain sharper and more actionable insights from the large swaths of data available. Cloud-based computing is more available than ever, which in turn has led to an increase in powerful and marketable risk products that can blend hundreds of otherwise disconnected indicators into a holistic patient risk profile.

While achieving true precision population health is not possible for many organizations today, it's important to note that organizations have already been building toward it.


A new tool to reduce costs and maximize return on investments

Organizations using precision population health can achieve two long-standing goals: reducing total cost of care through earlier intervention and maximizing the return on their population health investments.

  1. Reduction in total cost of care through earlier intervention: Covid-19 has led many patients to forgo preventive care and treatments, exacerbating existing conditions and increasing their risk of experiencing an acute episode. New studies also suggest that Covid-19 has led the average person to become more sedentary, consume more alcohol, exercise less, eat less healthily, and experience mental health needs. Precision population health can proactively identify rising patient complexity and risk, allowing providers to intervene before patients’ clinical and non-clinical needs culminate in costly and unnecessary utilization.
  2. Maximizing ROI through target interventions: In the Covid-19 recovery period, health care organizations are working to manage multiple workstreams (such as, Covid care, non-Covid care, contact tracing and testing, and vaccinations) amid an unsteady and uncertain financial future. Proactively identifying risk allows systems to invest their increasingly scarce resources into targeted solutions that yield the greatest benefit to the cohort's health outcomes and biggest cost reductions, maximizing the impact of their investments.

Early adopters

who's doing what

Application: Analytics-based cohorts for proactive socio-clinical interventions

In 2018, ProMedica partnered with Socially Determined, an analytics firm that measures the impact of social determinants of health (SDOH), to develop a precision population health model. Socially Determined developed a purpose-built social risk analytics platform, SocialScape®, which quantifies community-level SDOH risk exposure and individual-level social risk factors. Using AI, the platform integrates social risk data across multiple domains—such as food insecurity, housing instability, and transportation barriers—with historical clinical, claims, programmatic, and screening data. They use this information to proactively identify subsets of populations for which elevated social risk exposure is correlated with excess utilization, avoidable cost, or poor outcomes.

ProMedica then uses these insights to prioritize cohorts of patients with socially susceptible clinical conditions who are facing elevated social risk (such as patients diagnosed with congestive heart failure experiencing housing insecurity). The SocialScape platform quantifies excess costs associated with social risk exposure and prioritizes cohorts within ProMedica’s broader patient population for tailored outreach. Care managers then proactively match high-risk, high-cost patients within these cohorts to evidence-based clinical and non-clinical interventions that improve outcomes and reduce avoidable utilization and total cost of care.


What we’re keeping an eye out for

Data is more accessible than ever. As more organizations leverage their existing data repositories to develop their own precision population health models, we’ll be watching for advancements in three areas that could potentially make PPH models easier to implement:

  1. Critical data sets and combinations: New combinations of clinical and non-clinical data that can predict risk earlier (e.g., age and credit score).
  2. Automation: Tools that continuously and passively scan patient data from various sources, helping identify more patients with similar risk profiles.
  3. More powerful non-clinical interventions: Non-clinical interventions that may that yield a higher return on investment and are more beneficial to patients and total cost reduction than others.

The next generation of precision population health models may also be able to predict SDOH needs for patients who are not connected to the system, as ProMedica and Socially Determined's model does. Progressive systems may also leverage AI to proactively identify risk among patients in the community at large and intervene before those patients incur avoidable costs to the system.


“How ProMedica Prioritizes Data to Address Social Determinants of Health,” Radio Advisory, 24 September 2020; ProMedica; Advisory Board interviews and analysis; Bertrand L, “The impact of the coronavirus disease 2019 (COVID-19) pandemic on university students’ dietary intake, physical activity, and sedentary behavior,” Canadian Science Publishing, 15 January 2021; "Precision Population Health,” University of California San Francisco; Dolley S, “Big data’s role in precision public health,” NCBI, 7 March 2018;

Have a Question?


Ask our experts a question on any topic in health care by visiting our member portal, AskAdvisory.