1 Boring Old Man

Back in December, I tried to summarize the algorithmic treatment-by-computer obsession of Madhukar Travedi, a magnificent obsession…, cataloging the benign [IMPACT, an unfinished symphony], the expensive [STAR*D, another unfinished symphony], the negative [CO-MED], and the corrupt [TMAP]. 18 months ago, I had tried to schematize what he seemed to be trying to sell:

Feasible evidence-based strategies to manage depression in primary care.
^{by Kurian BT, Grannemann B, and Trivedi MH.}
Current Psychiatry Reports 2012 14[4]:370-5.

^{According to the World Health Organization, major depressive disorder [MDD] is a leading cause of disability-adjusted life years worldwide. However, recent evidence suggests depression remains undertreated in primary care settings. Measurement-based care [MBC] is an evidence-based strategy that can feasibly assist primary care physicians in managing patients with MDD. Utilizing health information technology tools, such as computer decision support software, can improve adherence to evidence-based treatment guidelines and MBC at the point of care.}

Adding another article to the growing list:

Here’s the conclusion to this most recent offering:

Conclusions:
In summary, recent clinical trial data from real-world practice settings reveal that the majority of patients will not achieve full remission of their depressive symptoms by the end of their first treatment and multiple treatment efforts are likely needed. To improve the likelihood of patients receiving adequate treatment and, accordingly, better outcomes, clinicians should apply feasible evidence based treatment strategies that involve systematic and objective monitoring of symptom severity, side-effect burden, and medication adherence. These three domains constitute the treatment construct known as MBC. In addition, MBC can also provide monitoring for residual symptoms associated with non-remission [insomnia, anxiety, suicidal behavior]. Specifically, as it relates to suicidal behavior, implementing patient-self reports, such as the CHRT and the CAST that assess for risk and associated symptoms, respectively, is an important tool for practitioners to monitor and assess for treatment-emergent suicidal ideation.

Furthermore, integrating MBC as part of a CDSS can maximize implementation- and utilization-recommended treatment guidelines, ensuring fidelity to up-to-date, evidence-based practices. However, prior to implementing a CDSS two items should be considered: [1] the identification of potential barriers to CDSS adherence; and [2] integration of the CDSS as part of the clinical workflow. For example, if a clinical practice is utilizing an EHR the CDSS should ideally be integrated as part of the EHR, such that double data entry is avoided. Similarly, depression, and associated mental disorders and residual symptoms rely on patient self-reports to guide clinical practice. As such, sufficient time should be built into a clinical visit such that patients are able to complete assessments prior to their face-to-face visit with the practitioner, and that the visit can then be utilized to jointly review treatment guidelines and decisions. In fact, when integrating a CDSS into clinical practice engaging the practitioners or end-users is an effective strategy to prevent barriers prior to implementation. Utilizing a shared decision model in which patients engage in the assessment and reporting of their depressive symptoms in primary care settings through a CDSS provides an opportunity for patient self-management and for guiding the dialogue for treatment visits, and it can improve adherence to evidence-based treatment guidelines and MBC at the point of care.

This all started with a hypothesis espoused by Drs. John Rush and Madhukar Trivedi at UT Southwestern in Dallas when it became clear that the then new antidepressants weren’t nearly so effective as originally hoped. They declared that it was because the doses weren’t brisk enough, the drugs hadn’t been tried long enough, the people who didn’t respond hadn’t gotten a trial of other drugs, there was a compliance problem, etc. The non-responders were said to have treatment-resistant depression, as if that were a specific disease. The underlying premise was that the antidepressants should work, and if they didn’t, somebody was doing something wrong. I think we can say with some degree of confidence based on their own studies that they were either wrong back then, or they were doing as many things wrong as the rest of us, because their results were abysmal [or absent].

So why does this set of ideas refuse to die? I guess it’s because if MDD were a unitary disease and if it had a biological substrate and if antidepressants were the specific treatment, then the issue would actually be one of effective treatment delivery.  Depression is certainly not a unitary disease. It is, by any report, a heterogeneous collection of a lot of things. Some depressions seem likely to have a biological underpinning, but many [or most] don’t. And there’s no real proof that the drugs are a specific treatment – only that they are empirically helpful in some patients. But this if³ hypothesis has great appeal to public health types wanting to standardize healthcare delivery systems. MBC, CDSS, EHR,  end-users, evidence-based, measurement-based, QIDS, CHRT, CAST, etc. All of those things play quite well in meetings, on Power-Point slides, and apparently on grant requests. People really, really want to objectify the subjective experience of depressed people.

It’s obvious from my comments that this is not my kind of psychiatry or mental health care. But the bigger point is that it was just a hypothesis of some Texans that now has a track record that should have retired the side a long time ago. Enough already. TMAP_TX,JNJ, CMAP_TX,JNJ, STAR*D_NIMH, IMPACT_NIMH, CO-MED_NIMH, EMBARC_NIMH.