Accountable care organizations (ACOs) were established to reduce costs and improve quality of care, but do they achieve those goals?
• A new study found that hospitals that participate in ACOs generally do not have lower readmission rates.
• Only patients with heart failure had a decrease in readmission rate in the ACO group.
• One factor could be that ACOs do not confer enough downside risk to health systems. With more financial risk, their care coordination programs resulting from the ACO model might work better.
The Affordable Care Act established accountable care organizations (ACOs) to reduce costs and improve quality of care, and some studies have shown positive outcomes. But new research examined the impact of hospitals’ ACO participation on their performance on readmission rates for several conditions. The results are surprisingly limited.
“The ACO model seeks to save money by care coordination and case management, but is it showing the benefit we’re looking for?” says Mark Diana, PhD, associate professor and chair of global health management and policy at the School of Public Health and Tropical Medicine at Tulane University in New Orleans.
Diana’s research suggests, “not so much.”
The study examined whether hospitals that participate in ACOs have lower readmission rates on average than those that did not participate — and what investigators found is mixed. “In general, they don’t. But with heart failure, there was a lower readmission rate,” Diana says.
The study found that ACO participation is significantly associated with a decrease in the heart failure readmission rate, but not in any change in readmissions among patients with acute myocardial infarction or pneumonia. Its conclusion is that “Medicare ACO programs have limited effects on readmission rates.”1
“There are a number of nuances. Fundamentally, the study we did was to look at hospitals participating in ACOs,” Diana says. “If you’re participating — even if it’s only for a Medicare population — then those activities also will impact other patients in the hospital, and we should see a reduction or improvement in outcomes metrics across the hospital.”
Research data did not separate the Medicare population from the general population with those same health conditions. “It could be if we examined just the Medicare population in each of those hospitals, we would see a difference,” he says. “But data was not provided that way.”
Also, future ACO programs might be set up with tracks that shift more downside risk to providers. It is possible that once hospitals stand to lose significant funds if their care management programs fail, they will put more resources into these programs and show better outcomes as a result, he adds.
“Maybe these ACOs don’t bear enough risk to move in that direction,” Diana says.
Another factor could be the type of patients helped in ACO programs. It is possible that benefits to a program are best when patients are most in need of help.
“What it shows depends on the severity of the illness of the patients in the population you’re looking at,” he says. “The more severely ill, the more likely they are to benefit from care coordination activities.”
Investigators found a similar pattern in patient satisfaction scores. There was not much effect in ACO hospitals versus non-ACO hospitals.
“I would have expected satisfaction scores to be higher,” Diana says.
The study does not suggest that case management and care coordination do not work: “When you have patients that are ill with comorbidities and complex health problems, they tend to benefit from care coordination/case management activities,” he says.
Alternately, healthy patients who regularly see their doctors are a population that would not benefit as much from case management and care coordination, he adds.
When health organizations participate in an ACO, it is expected they will increase case management and care coordination, but the study did not collect data on how many of the hospitals had developed such programs specifically because of the ACO. “We didn’t have the details on what kind of specific programs they were implementing,” he says.
“My primary interest is around how these new models of care are influencing cost and quality of care,” he adds. “What are these new care models, and do they have the potential to bend ... the cost curve and keep quality up?”
Diana also has researched how ACO participation influences hospitals’ performance. Another recent study found that ACO participation improved some aspects of patient experience in hospitals with prior good experience but did not benefit hospitals with historically poor performance.2
Patient experience data came from the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) scores. HCAHPS scores are used for nearly one-third of incentive-based payments from Medicare’s Value-Based Purchasing Program.2
The study found similar reported results between ACO hospitals and non-ACO hospitals on measures that include the following:
• communication with nurses;
• communication with doctors;
• responsiveness of hospital staff;
• pain management;
• communication about medications;
• cleanliness of hospital environment;
• discharge information;
• overall hospital rating;
• would recommend the hospital.2
One factor, quietness of hospital environment, had a higher rating among non-ACO hospitals than ACO hospitals — 61.70 vs. 57.76.2
The study concluded that hospital participation in ACOs does not lead to improvement in patient experience, but hospitals with a track record in care coordination are more likely to benefit from participation in an integrated care delivery model like that of ACOs.2
Diana also has researched the characteristics of hospitals that decided to participate in an ACO. “An interesting question in my mind of hospitals doing this kind of work before the ACO model came along is how they fare in the ACO model versus those not doing this work,” Diana says. “So I looked at hospital characteristics.”
1. Duggal R, Zhang Y, Diana ML. The association between hospital ACO participation and readmission rates. J Health Manag. 2018;63(5):e100-e114.
2. Diana ML, Zhang Y, Yeager VA, et al. The impact of accountable care organization participation on hospital patient experience. Health Care Man Rev. 2018: Aug. 3; Epub ahead of print.