Properly designed niche policies can have a tonne of positive spillover effects, suggests research by U of A professor Mohamad Soltani

Can a narrow public policy achieve broader results? That’s the question a team of researchers — including Mohamad Soltani, a business analytics professor at the Alberta School of Business — addressed in a paper recently published in the flagship journal Management Science.

The research studies the United States’ Hospital Readmissions Reduction Program (HRRP), a national healthcare policy that provides financial incentives for hospitals to reduce the number of patients — with specific clinical conditions and insurance type — who are readmitted to the hospital within a month of an initial hospital visit. 

The study reveals that the impact of hospitals’ readmission reduction efforts in response to the HRRP extended beyond the policy’s targeted demographic. As a result, the policy led to hundreds of millions of dollars in annual reductions in healthcare expenditures thanks to both its intended and spillover effects. 

“This highlights that policymakers should assess and account for potential spillover effects when designing partial incentive policies, particularly when the cost of administering a policy grows with its complexity,” Soltani says. 

Millions of dollars at stake

The HRRP is designed to reduce readmissions among a specific group: Medicare patients aged 65 and older with one of three clinical conditions — acute myocardial infarction, heart failure or pneumonia. (Soltani notes other conditions have since been added to its scope). 

To reduce readmissions for these patients, hospitals identified the underlying causes of readmissions and found solutions — such as enhancing adherence to medication guidelines and improving patient education for post-discharge care. These solutions — proved effective in reducing readmissions, Soltani says. 

Beyond the direct effect of the HRRP, this research delves into the effects of the policy on patients who were not targeted by it. “The impact of the accrued knowledge is not limited to target patients. Rather, shared resources between target and non-target patients, as well as communication among caregivers across the hospital, facilitate knowledge transfer,” Soltani says. 

This occurs because hospitals are typically organized by clinical specialty (e.g., cardiology, internal medicine), and achieving improved quality outcomes for a patient often requires collaboration among caregivers from multiple departments who serve both target and non-target patients.

The team of researchers used a quasi-experimental method, known as “difference-in-differences,” which estimates the causal effect of a policy or an intervention by comparing the outcomes for a treatment and a control group.

They used four years of data from the Nationwide Readmissions Database, created by the Agency for Healthcare Research and Quality (AHRQ). Each year in the database includes approximately 15 million observations, representing about half of all hospitalizations in the US, drawn from more than 2,000 hospitals from across the country.

The paper shows that, in addition to reducing readmissions for target patients, the HRRP led to significant quality improvement spillovers for non-target patients that have at least one common attribute (clinical condition or insurance) with target patients.  

According to the AHRQ, the average cost of a hospital readmission is around $16,300. As such, this study shows healthcare expenditures decreased by around $600 million annually thanks to the HRRP’s spillover effects, on top of the roughly $200 million per year in savings from targeted patients readmissions. 

Narrow focus, broad impact

While the paper focuses specifically on the HRRP, this study underscores the far-reaching effects of quality improvement policies, highlighting the potential for quality spillovers to amplify their benefits, Soltani says. 

He adds that this work more broadly offers novel insights into how the government and private sector can design narrow policies that achieve broader results by leveraging beneficial quality improvement spillovers.

Soltani is currently working on a follow-up project on the HRRP that examines another design aspect of the policy — the focus on 30-day readmissions, regardless of when the readmission occurs within that period.

In this ongoing project, Soltani and his colleagues aim to show that a one-size-fits-all policy, like the HRRP, can have varying impacts depending on the hospitals involved. This raises questions about fairness, he says, particularly considering the differences between institutions.This is especially concerning for how such policies disproportionately affect ‘safety net hospitals,’ which serve low-income patients with limited access to care, compared to other hospitals.