Examining the Impact of Behavioral Health Costs on Overall Health Care Spending

Of the top 10% considered high cost from an overall cohort of 21 million, a recent study by Milliman found that 57% of this high-cost group had a mental health or substance abuse diagnosis. This behavioral subgroup was shown to contribute to 44% of all health care spending, said Henry Harbin, MD, leading behavioral health expert and adviser to The Bowman Family Foundation.

Transcript

AJMC®: Hello, I'm Matthew Gavidia. Today on the MJH Life Sciences’ Medical World News, The American Journal of Managed Care® is pleased to welcome Dr Henry Harbin, a leading behavioral health expert who served as a key architect of a study by Milliman on how behavioral health conditions contribute to physical and total health care spending.

Recently, Milliman released a study, titled, “How do individuals with behavioral health conditions contribute to physical and total health care spending?” How was the study conducted, and were there any trends you were especially interested in learning more on?

Dr Harbin: The more recent study that I think you're referring to was really meant to look at the impact that people with these illnesses have on total medical costs. So, in this study, Milliman picked only the year 2017 to look at this data, the most recent year they have data in their databases and there they looked at 21 million people. These are, again, commercially insured populations, and what we did was we looked at, or they looked at the 10% most expensive patients mostly driven by physical medical costs.

So, we looked at that and what percent did that 10% have on driving or accounting for total medical spending, total health care spending. So, that 10% of the population drove, contributed, caused, whatever term you want to use, 70% of all health care spending. So, 10% of the 21 million people contributed 70% of the cost for the 21 million people.

Now, then, the purpose of the study was to look at the behavioral health subgroup, both in the high cost population and in the total 21 million lives that were measured–their claims. In the top 10%, 57% of that group had a mental health or substance abuse diagnosis, which is very high, much higher than any of us expected, and that subgroup of the 10% contributed 44% of all health care spending.

So, even though it was basically less than 6% of the total population, 21 million accounted for 44% of all health care spending for that population. Now, if you imagine that spread out over the whole health care population for all beneficiaries, Medicare, Medicaid, we would expect the Medicaid numbers would be even more dramatic than these numbers because there's typically a lot higher prevalence of serious and moderately ill mental health substance use problems in Medicaid, just by the nature of the insurance.

So, that would basically mean if you look at the whole US health care–over 3 and a half trillion dollars a year spending, basically 6% or 7% of the population that have 1 of these comorbid illnesses are going to drive almost half of all those things.

AJMC®: Based on study findings, a significant gap was represented between those suffering from behavioral conditions and their ability to afford treatment, with 50% of all those with these conditions having less than $68 of total annual spending. What factors contribute to this gap and how can these lack of coverage options impact overall health and health care spending?

Dr Harbin: So, the other finding that really wasn't the focus of this study, it wasn't a study on disparities as much as a study to look at the outsized impact that a mental health condition might have on medical spending, was how little specialty care these people got. So, for the whole population of the behavioral group who had a behavioral diagnosis or got treated for a behavioral problem–50% of the group got less than $68 a year of specialty behavioral treatment, and about 25% got less than $500–somewhere between $68 and $500.

So, essentially, 75% of this population got almost none or very little specialty behavioral treatment, despite them being diagnosed by a licensed health care professional who felt they warranted a diagnosis and warranted treatment. So, that was a bit of a surprise–as I mentioned these 2 earlier Milliman studies showed a lot of disparities in access, but we didn't measure the percent of the population, how little or how much care they got.

So, those 2 findings that were unusual, not been reported much, there's been, as you probably know, a lot of data show that a small percent of chronic medically ill patients drive a big impact of total health care costs, but there's been less attention paid to what percent of that group has mental health and substance abuse problems.

As you look at the report that Milliman did, it's interesting that the majority of these costs are driven by people with mild to moderate mental health, substance abuse problems. There's commonly a misconception it's the patients with more severe mental illnesses like psychosis or schizophrenia that are the ones that drive these costs. That's not true. They represent a very small percent of prevalence, particularly for the commercial population, but even for other insurance beneficiaries.

So, those are people that probably could be the easiest to treat, and yet weren't–they weren’t getting that treatment. So, I think the main takeaway for health care policy leaders on the provider side or the health plan side or employers is that if you want to have an impact on total medical spending, you better identify this group early and provide some specialty intervention sooner.