This month's Journal of Clinical Psychiatry has a lead article about medication guidelines for adults with major depressive disorder. Is is an apparent function of the Florida Medicaid Drug Therapy Management Program For Behavioral Health. It is hard to imagine a title with more inappropriate terms for what psychiatrists do with medications. At least until I read the title of the article: "Florida Best Practice Psychotherapeutic Medication Guidelines (FPG) for Adults with Major Depressive Disorder." Here is a little insight into what I have difficulty with. Treatment with medications is not psychotherapy. Psychotherapy almost always needs to accompany medication treatment at one level of intensity or another. But providing medication alone is not psychotherapy. That is an important distinction because one of the common misconceptions is that a medication will solve common life problems like interpersonal problems at work or home and it will not. The second issue is the idea of medication "management". As one of my colleagues used to say: "Pharmacists manage medications we treat patients". The term should also be anathema to any psychiatrist who was around when billing and coding guidelines were invented. The term came to mean 10-15 minute appointment that reduced psychiatric treatment to a brief discussion of a medication. They were two of the lowest reimbursement codes in the coding scheme and they handily allowed psychiatric treatment to be split off from the rest of medicine and reimbursed at a lower rate. And finally the term behavioral health. This is a long standing business term to indicate a managed care environment with business supervision rather than a mental health environment with psychiatric supervision. All of these terms suggest that managed care companies and the government have more to do with these guidelines than psychiatrists.
Sure enough, looking at the partners for this project the majority are behavioral health organizations or managed care companies followed closely by government organizations, other associations, and three psychiatry departments out of 24 organizations. The article itself describes the process as being the result of a multistakeholder Florida Expert Panel. The stakeholder word always makes me cringe. Whenever I have seen it in medicine and psychiatry nothing good has ever come of it. There are only two stakeholders in medical treatment - the physicians and the patient. I can stretch that to the family if they are still actively involved. I don't want to see anybody else in the room.
The authors detail their rationale for yet another guideline. They state:
"Notwithstanding the public health priority of MDD, as well as increasing public, academic. and policy attention given to MDD, misdiagnosis or delayed diagnosis and failure to incorporate appropriate measurement based care are significant modifiable deficiencies in current practice."
If only that were true. In a state where there is widespread PHQ-9 screening. the screening tool suddenly becomes the diagnosis. Measurement based care suddenly becomes the collection of meaningless cross sectional scores from clinics all over the state listing a diagnosis of MDD. If only real life worked like intensive clinical trials out to prove a hypothesis.
They go on to list several other reason for their guideline. They cite the American Psychiatric Association (APA) guideline as a "conflation" of empirical evidence and expert consensus - suggesting that nothing is sacred about expert consensus and that the patients seen by experts may not be the same as patients seen by other physicians. They suggest that guidelines derived from pharmacological trials may be limited by suggesting that they may have limited generalizability due to trail designs and conditions that rule out certain conditions, but don't discuss other problems in experimental design. They discuss limited long term follow up and measurement of functional capacity as a limiting factors. Given that the authors don't really intend to correct any of these criticisms it is difficult to see that as a rationale for the new guideline. Instead they say that their consensus process was their overarching principle in writing the FPG along with providing guidance (especially to primary care physicians) to provide safe and effective treatment for depression.
The authors use a hierarchical approach to tiers of treatment without using an algorithm. Level 1 is initial treatment and Levels 2, 3, and 4 are basically used if the initial levels of treatment are ineffective or not tolerated. There are few surprises for any psychiatrist who is used to treating depression, especially referrals from primary care physicians. Given the stated concerns about the biasing effects of clinical trials sponsored by pharmaceutical companies for specific FDA indications, there were not many qualifiers about the addition of an "atypical antipsychotic approved for major depressive disorder (ie. aripriprazole, brexpiprazole)" at Level 2. Level 2 is basically a failure of Level 1 antidepressant monotherapy. In fairness switching to another antidepressant monotherapy at Level 2 is a suggested option. The clear concern that the authors have about second generation antipsychotics in their scheme is metabolic rather than neurological side effects. I have found a significant number of neurological side effects from aripiprazole including Parkinson's syndrome, akathisia. and tardive dyskinesia from these medications. Nowhere in the paper are the diagnostic skills listed as important for the physicians. In the emphasis about measurement based care there are no rating scales for drug induced neurological disorders. The question of safer augmentation strategies are not discussed.
With regard to the issue of weight gain as a medication side effect, a strategy listed is "select medications that have a low relative risk of weight gain and metabolic syndrome". A couple of related issues come up including the fact that a significant number of patients walk into the clinic with high BMI, but they are there for the treatment of depression. Should the diagnosis of obesity and/or metabolic syndrome be made and managed along with the depression? What about the patients who gain significant weight on either aripiprazole or brexpiprazole? They definitely exist. What about clinicians who have developed successful strategies for using atypical antipsychotics with minimal to no weight gain?
There are also the very common problems of insomnia associated with depression that does not resolve with antidepressant therapy and significant anxiety with or without panic attacks. Major depression with psychotic features and major depression with mixed features were discussed as important variants and special interventions not commonly used in primary care were included like electroconvulsive therapy (ECT) or transcranial magnetic stimulation (TMS). It was acknowledged that lack of patient acceptance and availability of these treatments might result in using various medication combinations that may be less effective. Vagal nerve stimulation was recommended as a level 4 treatment and I have serious reservations about that being effective for anyone.
All in all the FPG is what I would expect from a collection of stakeholders, some of whom were listed as representatives of managed care companies. Rather than have these stakeholders rehash strategies that have been around for 20 years, there was an opportunity to design a comprehensive system of care for patients with depression and there is no evidence that has happened. There is a reason why people don't go in to psychiatry and some of those reasons don't bode well for the assumption that everyone in the system will now be doing comprehensive assessments like psychiatrists. A system of mental health care designed by stakeholders could possibly develop state- of-the-art resources for neuromodulation (TMS, ECT, deep brain stimulation), sleep studies, monitoring the cognitive effects of depression and antidepressants, detoxification and addiction treatment, and reasonable inpatient and residential resources. That same system would have designed in timely assessments of difficult problems like MDD with psychosis by psychiatrists. Adequate numbers of psychotherapists or pilot programs looking at computerized cognitive behavior therapy for sleep, depression, and anxiety would be more useful that one or two crisis oriented sessions with no specific orientation. A blanket statement about the utility of evidence-based psychotherapies without adequate numbers of therapists to carry it out is not helpful in any way.
We need system redesign by stakeholders, not stakeholders making more guidelines while pretending that they know something about quality.
George Dawson, MD, DFAPA
Synopsis:
If certain stakeholders in a system, have:
-all of the money
-all of the power
-sophisticated electronic health records that are set up more for administrative than clinical purposes.
They may have an obligation to design the system to optimize care rather than telling the people delivering the care what they can do in a poorly integrated system of rationed resources by applying strategies that are already well known.
References:
2: Ostacher MJ, Tandon R, Suppes T. Florida Best Practice Psychotherapeutic Medication Guidelines for Adults With Bipolar Disorder: A Novel, Practical, Patient-Centered Guide for Clinicians. J Clin Psychiatry. 2016 Jul;77(7):920-6. doi: 10.4088/JCP.15cs09841. PubMed PMID: 26580001.
3: Gartlehner G, Gaynes BN, Amick HR, Asher GN, Morgan LC, Coker-Schwimmer E, et al. Comparative Benefits and Harms of Antidepressant, Psychological, Complementary, and Exercise Treatments for Major Depression: An Evidence Report for a Clinical Practice Guideline From the American College of Physicians. Ann Intern Med. 2016;164:331-341. doi: 10.7326/M15-1813.
4: Qaseem A, Barry MJ, Kansagara D, for the Clinical Guidelines Committee of the American College of Physicians. Nonpharmacologic Versus Pharmacologic Treatment of Adult Patients With Major Depressive Disorder: A Clinical Practice Guideline From the American College of Physicians. Ann Intern Med. 2016;164:350-359. doi: 10.7326/M15-2570.
