It is well established that in the United States as a whole, and in various payer or demographic subpopulations, a relatively few patients account for much of total health spending. But it also appears to be the case that most patients who are high spending in one year will not be in the next year. A study in Health Affairs analyses one group of these so-called super-utilizers. (HA Article) This sub-group frequently has multiple chronic conditions and also is likely to have a number of social issues, from lack of housing or jobs to poor support systems. This research looked at primarily low-income patients in Denver served by one safety-net integrated provider. The authors looked for people with large numbers of emergency room and avoidable hospitalization visits. Even within their total identified population of super-utilizers, there was a subgroup of 3% that accounted for 30% of spending. Over the two-year study period, however, there was little persistence of super-utilizer status. Within seven months after the start of the study period only 28% of people who were super-utilizers in the first month retained that status. Some died, but for most their utilization and spending dropped. Spending in the first year for the original cohort averaged $113,522, but that average had dropped 60% two years later.
As this article suggests, from a care and cost management perspective the biggest problem is identifying in advance who is going to be a high-spending patient in a particular year, which all the predictive analytics in the world don’t seem to be capable of doing, and even once you know who they are in that year, the majority of your high-spending patients in that year won’t be in the same category the next year. The group that does have persistently high spending for several years is obviously worth focusing intense attention on and probably has a payback for the cost of those efforts. But finding in advance and managing the one-year high spenders is a problem that has not yet been solved and may not be. The best approach may be to understand as quickly as possible when a likely high-spending episode has been triggered and to immediately bring care management techniques into play. This research did identify distinct sub-types within the super-utilizer population; sub-types which are amenable to specific interventions for their needs.