Study Dissects Camden Coalition Results to Pinpoint Disconnect Between Care Coordination and Reduced Readmissions

The Camden Coalition’s “hotspotting” program, which delivered intensive care coordination services to high-risk, high-cost patients, was considered one of the most promising approaches to managing spending and improving care quality—until null findings from a randomized controlled trial showed no difference in readmission rates between patients in the program and those who received usual care.¹ Now, 4 years later, researchers including some authors on the original study have published new research that aims to explain why the care management program did not achieve its goal.²

The journey to this research began over 10 years ago, picking up steam when an influential article in the New Yorker by Atul Gawande, MD, highlighted the work of Jeffrey Brenner, MD, then the CEO of the Camden Coalition of Healthcare Providers.³ Camden, an economically challenged city in New Jersey, was struggling with its performance on health care quality metrics, and Brenner and his team created an intervention in which the most complex and costly patients would be targeted for intensive care management and connection to services that can help with social needs, such as transportation and housing.⁴ A team of social workers, clinical staff, and community health workers collaborated to encourage patients to follow their discharge plans, connect with a primary care provider, and address social needs, all with the goal of reducing readmissions to the hospital.

Then, findings published in the New England Journal of Medicine in January 2020 poured cold water on the hotspotting model.¹ Among 800 Camden Coalition patients with high health care utilization and complex social needs who were randomly assigned to either the care management program or usual care, readmission rates at 180 days after discharge were not significantly lower among those in the program. Secondary outcomes such as number of readmissions and different time horizons for readmission also did not show an advantage for the hotspotting group.

Four years later, several of the authors of that 2020 article have published in Health Affairs new findings examining how exactly the Camden Coalition model failed to have the expected impact.² Either the hypothesis that care coordination would prevent readmission was wrong or the program was not implemented successfully, the authors wrote, so they linked the original study participants to Medicaid data to test these potential explanations.

Their analysis was restricted to the 561 patients who were enrolled in Medicaid at the time of enrolling in the trial; 281 were in the treatment group and 280 were in the control group. In addition to the Medicaid claims data, the authors incorporated data from a baseline survey of demographic characteristics, hospital discharge data from the original trial, and billing data from the New Jersey Department of Health. With these data sources, the investigators were able to analyze the effect of the program on emergency department (ED) use and access to ambulatory office-based care.

Confirming the results of the original trial, the new analysis of the Medicaid sample found no effect of the program on 180-day readmission rates (P = .40). Using the ED data not available at the time of the original trial, the authors also found no differences in ambulatory ED visits 180 days after discharge, either in the full trial sample (P = .48) or the Medicaid sample (P = .75). However, in the analysis of ambulatory outcomes for the Medicaid sample, the treatment group was more likely than the control group to have had an ambulatory office visit within 14 days after discharge (42.35% vs 27.14%; P < .001), including any primary care visits (33.10% vs 18.93%; P < .001). The numerical gaps between the groups narrowed by 180 and 365 days after discharge, but the differences remained statistically significant. For instance, by 180 days after discharge, 83.63% of the treatment group had an ambulatory office visit, compared with just 73.21% of the control group (P = .002).

Although not significant at 14 days, the treatment group showed higher rates of specialist visits than the control group, and the differences were significant at 180 days after discharge (66.55% vs 52.5%; P = .001) and 365 days after discharge (77.94% vs 66.43%; P = .002). The program was also associated with an increased likelihood of receiving durable medical equipment, but no effect was found on the number of prescription medications that patients received.

These observed increases in ambulatory care are similar in magnitude to those found by other studies of care coordination programs, the authors noted, leading them to conclude that although implementation of the Camden Coalition hotspotting model was not perfect, it did improve patients’ likelihood of visiting a care provider after discharge. However, increasing these postdischarge visits was not enough to prevent readmissions.

“We therefore interpret our findings as consistent with the view that care coordination alone, even when implemented well, is insufficient to reduce hospitalizations for this complex population of high-need patients,” the authors wrote.

Next steps for the evaluation of care management models will include greater incorporation of nonclinical supports, such as housing and legal services, to address the nonmedical drivers of poor health outcomes. Randomized trials of these efforts will shed more light on their potential to control costs and improved outcomes, the authors concluded, but even null results from such trials can yield important insights, as in this fresh look at the original Camden Coalition findings.

References

1. Finkelstein A, Zhou A, Taubman S, Doyle J. Health care hotspotting - a randomized, controlled trial. N Engl J Med. 2020;382(2):152-162. doi:10.1056/NEJMsa1906848

2. Finkelstein A, Cantor JC, Gubb J, et al. The Camden Coalition care management program improved intermediate care coordination: a randomized controlled trial. Health Aff (Millwood). 2024;43(1):131-139. doi:10.1377/hlthaff.2023.01151

3. Gawande A. The hot spotters. New Yorker. January 16, 2011. Accessed January 9, 2024. https://www.newyorker.com/magazine/2011/01/24/the-hot-spotters

4. Joszt L. Hotspotting: care intervention for the most complex, costly patients. AJMC. October 21, 2016. Accessed January 9, 2024. https://www.ajmc.com/view/hotspotting-care-intervention-for-the-most-complex-costly-patients