During the pandemic, it became clear that a "one-size-fits-all" approach was not effective at protecting or improving the health of the population. Going forward, precision population health (PPH), a more tailored approach to healthcare, will be needed to proactively prevent and treat health conditions.

What is precision population health?

PPH combines the tailored approach of precision medicine with the broad scale of population health.

PPH accounts for people's differences in genes, lifestyles, and environment. It uses a predictive form of risk stratification to identify the specific healthcare needs in a community and offer interventions to target those specific needs.

Rather than reacting to health problems that already exist, PPH tries to proactively identify patients' potential health needs and connect them with both clinical and non-clinical interventions to help prevent high-risk and high-cost health conditions.

Compared to traditional population health, PPH targets a narrower cross-section of individuals. While traditional risk-stratification frameworks place patients into low-, moderate-, and high-risk categories based on disease burden, utilization, and cost, PPH uses large data sets to more precisely calculate risk and group patients into specific cohorts.

An early adopter of PPH was ProMedica. In 2018, ProMedica partnered with the analytics firm Socially Determined to create a social risks analytics platform, which used artificial intelligence (AI) to analyze social determinants of health (SDOH) with historical clinical, claims, and screening data. The platform also quantified the excess costs associated with exposure to social risks, such as food insecurity and housing instability.

These analyses were used to prioritize cohorts of patients who had socially susceptible clinical conditions and elevated social risks. Care managers then matched high-risk, high-cost patients to evidence-based care to help them improve their health and reduce unnecessary treatments and costs.

The future of PPH

Over the last 10 years, large data repositories have become easier and cheaper to access, which has allowed organizations to incorporate more nontraditional data sets (credit scores, purchasing habits, and more) into their risk analyses.

Recent advancements in computing power have also allowed organizations to analyze the data they have more quickly and thoroughly, which has led to sharper, actionable insights. These analyses can combine hundreds of data points into a holistic patient risk profile.

In the future, advancements in three areas could help health systems implement PPH models more easily:

1. Critical data sets and combinations: New combinations of clinical and non-clinical data, such as age and credit score, may help organizations predict patient risks earlier
2. Automation: Tools that automatically scan various sources for patient data can help identify more patients with similar risk profiles
3. More effective non-clinical interventions: New non-clinical interventions may have a higher return on investment, be more beneficial to patients, and reduce total costs most effectively

In addition, new PPH models may be able to predict SDOH needs for patients who are not connected to a particular health system. Some health systems may also use AI to proactively identify risk in communities and provide interventions to high-risk patients before they lead to unnecessary treatments and high costs. (Lee, Forbes, 5/12; "Precision Population Health," Advisory Board, 3/17/21)