There is a perception that one class of frequent inappropriate ER utilizers are the mentally ill and substance abusers and that these people, who often are covered by Medicaid, are imposing a significant unnecessary cost on the system. Research in Health Affairs suggests otherwise. (HA Article) The authors used data on Medicaid recipients in New York City to examine characteristics of ER users. The definition of a frequent ER user is unclear, but those who have 3 to 5 visits to an ER a year tend to have multiple chronic conditions, frequent hospital admissions and high use of primary and specialty care, so they likely are really sick. While the uninsured and Medicaid have high rates of ER use, the majority of frequent users are caucasian and covered by Medicare or private insurance. Only a small number of users, 1.7%, have ten or more annual visits, but 29% of users have three or more, accounting for 60% of all visits. High users tended to be disabled and really high users were less likely to be in managed care. Fifty percent of all ER users had a chronic illness and among very high utilizers, 85% had at least one chronic disease. High users also had significant rates of substance abuse and mental illness, with utilization increasing with rates of these problems. Heavy users did not always go to the same ER, very heavy users on average went to five. In terms of the reason for the visit, substance abuse and mental illness were fairly low at around 5% each, although for the greatest users, these reasons were also higher. The heavy users, as noted above, tend to use a lot of all kinds of care, so lack of access is not likely to be a reason for using the ER. The authors also did some predictive modeling, finding that heavy use of the ER in one year was very predictive of heavy use in the next, as was the presence of multiple chronic conditions, a mental illness or substance abuse diagnosis and high use of primary care services. What the report overall suggests is that these patients who are high utilizers of the ER have significant health problems and health care service use and therefore are good targets for comprehensive interventions to manage their overall health and care. One thing to be aware of is that social circumstances, such as living arrangements, can be critical in ER use, and this study did not have access to that data, but any successful intervention likely needs to address those social needs..
