We sail into uncharted waters... GSC tackles mental health

June 24, 2016

GSC’s The Inside Story has covered a lot of ground over the years. But one topic has been conspicuously absent—mental health. In the benefits landscape, this issue has been linked mostly to disability management. Canadian plan sponsors’ short- and long-term disability portfolios are full of claims with origins that lie in depression and anxiety disorders. If it is not the leading cause of disability claims and costs in your organization, it is probably number two.

So, what possessed a health and dental benefits specialist to turn its attention to mental health? Although we’ve been silent externally on this issue, we were not quiet inside our four walls. We have been examining drug claims data, consuming research, and debating among ourselves whether there is a fresh perspective we can bring to this critical topic

Charting our course

The 2015 GSC Health Study is our coming out party. We constructed a chapter entitled “It’s All in Our Heads”; not to suggest that mental health issues are imaginary, but to acknowledge the amazing complexity of the human mind, the challenges of accurate diagnosis, and the emerging science related to proper prescribing of antidepressant medication.

Our all-consuming fascination with data is a matter of public record, and our recent examination of antidepressant claims patterns may be the most intense and in-depth study of drug claims we have done yet. Starting last fall, we worked with our data-obsessed friends at Cubic Health Inc. to look at reams of GSC data related to the claiming of antidepressants.

Before the big “unveil” of our analysis, let’s share our perspective going into the Health Study:

  • The Journal of the American Medical Association (JAMA) published an expansive study in 2010 that suggested the newest generation of antidepressants better served a population of patients with severe depression rather than mild to moderate depression. This suggested that “the system” is likely over-prescribing these drugs to the mild to moderate depression population.
  • The study also demonstrated that the mild to moderate depression population is being prescribed very low-level dosages, “sub-therapeutic” to be exact.1
  • In the broader Canadian health care environment, there is a lack of evidence-based, quality-controlled cognitive behavioural counselling available in our communities, removing from physicians a critical and effective option that is considered a first-line treatment for depression, either alone or in conjunction with medication.2

Our mission, which we accepted, was to determine whether our prescription drug claims data would confirm or deny these theories

Antidepressant Drug Effects and Depression Severity: A Patient-Level Meta-Analysis

This paper published in JAMA in 2010 reported the results of a study that assessed the effectiveness of antidepressants by analyzing existing drug trial data.

Findings:

“The magnitude of benefit of antidepressant medication compared with placebo increases with severity of depression symptoms, and may be minimal or nonexistent, on average, in patients with mild or moderate symptoms. For patients with very severe depression, the benefit of medications over placebo is substantial.”3

In other words, antidepressants were minimally effective, if at all, for people with mild to moderate depression; however, antidepressants were much more effective in people with very severe depression.

About the study:

  • It analyzed data from six randomized placebo-controlled trials of antidepressants approved by the U.S. Food and Drug Administration.
  • Each trial included a medication versus placebo comparison for at least six weeks.
  • A total of 718 adults were studied; they had been diagnosed with minor to major depressive disorder.
  • Three of the trials were for a selective serotonin uptake inhibitor (SSRI) and the other three trials were for a tricyclic medication (an older type of antidepressant).4
Buckle up, here are the GSC numbers...

Step 1: We identified 350,000 GSC plan members who, over a three-year study period, had continuous prescription drug coverage and experienced no change in plan design. Yes, a “clean” sample for doing our assessment. 

Step 2: In that three-year study period we identified 35,000 of those plan members who started on an antidepressant (we call them “new starts”). Those are the folks we zeroed in on to determine what we could learn from the prescribing and claiming patterns we had in our data. 

So let us tell you the story of the 35,000 “new starts.” We divided this large group into two—one group we called Treatment Resistant Depression (TRD), the other group we called Monotherapy. (Look at us, sounding all sciencey!)

THE TRD GROUP

Let’s talk about the TRD group. What jumps out in the claims data first and foremost is that these plan members try more than one antidepressant, and we see changes in dosages over the course of their claims history; this suggests robust follow-up. What is evident is a physician trying to determine the best drug regimen for a complex and very sick patient. And that’s good to see— clear efforts to treat what we believe are the severely depressed population cited in the JAMA study above.

But, take note, this group makes up only 12 to 15 per cent of those 35,000 new starts on antidepressants in the three-year study period—a distinct minority. If there are headlines to be written from our mental health analysis, they do not lie here. We believe the evidence in our data suggests this group is getting adequate care. But there are many more question marks related to our Monotherapy group…

More on the TRD group:

  • Average age is 45.
  • They are high-cost claimants, definitely part of the top five per cent of plan members that account for 50 per cent of GSC drug expenditures (see the May issue of The Inside Story for more).
  • Only 35 per cent of their drug spend is for antidepressants, 65 per cent is for drugs related to hypertension, cholesterol, and diabetes—a roll call of chronic disease.
  • They have the highest rates of adherence to antidepressants that we have ever identified—over 60 per cent—nothing to write home about, but materially higher than the 45 per cent adherence we see in the general GSC antidepressant population.

The Monotherapy group

This group comprised 85 to 88 per cent of the 35,000 new starts on antidepressants in the three-year study period. This is the vast majority of our study group. Those of you who remember studying ancient Greek in high school know that “mono” means single or one (and if you took ancient Greek, it means you’re old). And our Monotherapy group had a clear profile in the GSC claims data. This group of plan members is prescribed a lone antidepressant, frequently at a sub-therapeutic dose, and a large proportion of them drop off the medication soon after therapy starts. This is a very concerning pattern. Very.

Time to dive in…

First let’s define what “sub-therapeutic” means, because that word is in our potential headlines from the study. There are clear guidelines for physicians on prescribing medications. In those guidelines, it states the levels at which the drug starts to have an impact on humans. We saw prescriptions below that level in 44 per cent of our plan members starting on antidepressants. (It’s appropriate for our readers to mouth the word “wow” at that number.) The rest of the group looks like this:

  • 50 per cent of the monotherapy group received a starting dose at the minimum therapeutic level.
  • Six per cent were prescribed above the minimum therapeutic level.

Earlier we suggested that a substantial number of the Monotherapy plan members dropped off their antidepressants rapidly— specifically within the first 135 days. This data must be presented with the knowledge that it is generally recommended that a patient continue on an antidepressant for six months following the resolution of symptoms.5

On that front, 26 per cent of the plan members receiving a sub-therapeutic first dose did not fill a second prescription. Therefore, they were likely on the medications for 30 days and then off. And 68 per cent were off their medications before 135 days passed. Of the group receiving the minimum therapeutic dose, 25 per cent did not have second fill, and 20 per cent of the above therapeutic dose group never filled twice.

So what are the numbers telling us?

We started this article by recognizing the complexity of treating mental health. As we have presented this data across Canada over the spring, we have been careful around the conclusions we draw. There are no CT scans, MRIs, or blood tests for physicians to rely on to help diagnose mental health. And the prescribing of antidepressants is less predictable in its results than prescribing a statin for hypertension or cholesterol. So we have been careful to present the numbers in a way that doesn’t lead to a position that doctors, likely family physicians in the majority of these cases, are failing to properly treat patients with depression and/or anxiety.

Let’s try to define some fair and balanced learnings…

We see a mental health eco-system where family physicians do not have a robust toolkit. Is the prescribing of a sub-therapeutic dose a natural outcome when a physician is faced with a patient needing support—a patient who is not severely depressed or anxious, but going through a temporary period when support and treatment is needed? The physician’s first instinct may be to refer a patient to cognitive behavioural counselling but the truth is, it is simply not widely available. And we as patients, in our modern world, expect our doctors to do something. To that end, a sub-therapeutic dose prescribed by that physician is not likely harmful to the patient, but our concern from the study is that it is not helpful. We have now had the opportunity to share these results with physicians—both family doctors and psychiatrists—and they agree that this is the dilemma physicians face in their day-to-day practice.

Our readers know we have a history of mining our data to identify the costs that drive our health benefits plans. We have also done our best to take the next step and try to determine whether concrete health outcomes result from that investment in health benefits.

So, yes, the headline here is that we are investing a lot of dollars in antidepressants that are not having a downstream impact on plan member health. GSC reimbursed $45 million in antidepressants last year. A significant portion of that spend reflected the prescribing patterns demonstrated in this new data.

Back to our talk about a physician’s toolkit. Imagine a system where evidence-based, quality-controlled cognitive behavioural therapy did exist for Canadians. And the dollars currently spent on un-impactful drug therapy were invested in that. That is what research tells us will be more effective for portions of the mild to moderate depression population.

WHAT NOW?

At a time in our country when we have highly impactful campaigns telling people to put up their hands and seek out care for mental health issues, it would be terrific if the care they were met with was far more effective.

Our next step...

Is to try to figure out the path to get to that better place—and here is a warning: Now that we have broken our silence on mental health, you may not be able to shut us up.