Remind Me Why Managed Care Companies Don't Offer Computerized Psychotherapy?

One of my mentors in residency enlightened me to the advantages of computerized psychotherapy.  That was in the days of the IBM AT PC or about 35 years ago.  I had an updated conversation with him about this 4 years ago after he gave a presentation that examined the more recent literature.  That literature had not only taken off, but there were more widespread applications.  Many of those applications were on a large scale using standard psychotherapeutic approaches.  Some of the research addressed the issue of patient acceptance and found that it was generally acceptable and more convenient for patients.

Interestingly, these innovations in computerized psychotherapy occur at a time when healthcare companies in the USA are implementing psychopharmacology on a large scale with minimal input from psychiatrists.  The collaborative care model for depression consists of patient in primary care settings rating depression with a standardized rating scale and generally being put on an antidepressant.  There is much wrong with this model, not the least of which is the fact that a rating scale is not a diagnosis and it probably leads to broader exposure to antidepressants medications.  It is nonetheless heavily promoted as an acceptable approach to the treatment of depression.

In the meantime, large organizations like the National Health Service in the United Kingdom have computerized psychotherapy services for the treatment of depression, generalized anxiety, and panic disorder.  The evidence base for these services is provided by the National Institute and Care Excellence (NICE).  According to the NICE guidelines for depression, computerized cognitive behavioral therapy is recommended as  "step 2 services for people with subthreshold, mild and moderate common mental health disorders".  In the US there are no similar resources and the treatment of depression as identified by screening techniques resembles a model of active monitoring and pharmacology with the widespread use of antidepressants.  Psychotherapy within health care systems in the US is sporadic and when it does occur seems to be based on a model of crisis counseling rather than any research model of the psychotherapeutic treatment of a specific disorder.  A main driver of that delivery system is the very low expense of modern generic antidepressants.  Some of these medications coast as little as $4/ month and are available at another discount if 90 day supplies are ordered.

A recent review of NSDUH data looked at medication utilization for the treatment of mental illnesses and found that the overall prescription rates had increased substantially across different generations of outpatients.  According to the authors there was a increase in prescription of psychiatric medications for all adults of about 28% between 2008 and 2013.  Of course the bulk of that medication is not prescribed by psychiatrists.  That paper also referenced a meta-analysis of patient preference for psychotherapy (2).  That study has significant methodological limitations and typical clinical constraints are probably not well represented.  In clinical practice, severely ill patients with psychiatric disorders or addictions are typically seen in intensive setting where psychotherapy is available and in many cases necessary like DBT or GPM.  The patient's I see list lack of rapport with the therapist, lack of direction in therapy, excessive self disclosure on the part of the therapist, cost, and inconvenience as being the main limiting factors for psychotherapy.  Most if not all of these constraints disappear with a computerized approach.

That brings me to a good example that I mentioned in a recent post on mindfulness based approaches to psychotherapy (3).  In the article I referenced, the Headspace web site was mentioned as a mindfulness based approach to psychotherapy.  The site offers 10 - 10 minute sessions and a more extensive selection of sessions at a reasonable cost after that.  As I completed the first 10 sessions, the advantages were immediately obvious.  The therapy model is obvious and there are standardized metaphors as examples.  The therapy is consistently delivered in the comfort of your own home and on your own schedule.  The site can be used by any psychiatrist or psychotherapist to augment outpatient treatment.  The site is valuable to trainees who are attempting to learn and deliver these mindfulness interventions on their own.  The site is cost effective - you can keep repeating the initial 10 sessions for free as much as you want or purchase additional sessions with different content.  A lifetime subscription to the web site is available for what might be the out-of-pocket cost of two psychoanalytic/psychodynamic sessions.

Every time I have been on the web site there are about 40,000 users logged in to the site.

Every managed care and health care company has a massive investment in IT to support their electronic health record and other clinic services.  Offering computerized psychotherapy would be a minor addition to those services.

Remind me again why managed care companies are not offering this option for depression, anxiety, and insomnia?    

And remind me why this is not the first step before prescription medication, especially in primary care settings?



George Dawson, MD, DFAPA



References:

1:  Han B, Compton WM, Mojtabai R, Colpe L, Hughes A. Trends in Receipt of Mental Health Treatments Among Adults in the United States, 2008-2013. J Clin Psychiatry. 2016 Oct;77(10):1365-1371. doi: 10.4088/JCP.15m09982. PubMed PMID: 27486895.

2: McHugh RK, Whitton SW, Peckham AD, Welge JA, Otto MW. Patient preference for psychological vs pharmacologic treatment of psychiatric disorders: a meta-analytic review. J Clin Psychiatry. 2013 Jun;74(6):595-602. doi: 10.4088/JCP.12r07757. Review. PubMed PMID: 23842011; PubMed Central PMCID: PMC4156137.

3:  Smallwood J, Mrazek MD, Schooler JW. Medicine for the wandering mind: mindwandering in medical practice. Med Educ. 2011 Nov;45(11):1072-80. doi:
10.1111/j.1365-2923.2011.04074.x. PubMed PMID: 21988623. http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2923.2011.04074.x/full

Within the body of that article the web site called Headspace ( https://www.headspace.com/ ) for mindfulness training is referenced.

Disclosure:

I have no conflict-of-interest in mentioning the Headspace site.  I am not endorsing it and they do not endorse me.  I have no financial interest in the site.  I am listing it here as a resource for mindfulness based approaches because I think it is useful and it was recommended in the literature by a psychologist who I consider to be an expert in this field.

University of Wisconsin 3rd Annual Update

I just finished the 3rd Annual Update and Advances in Psychiatry at the University of Wisconsin in Madison Wisconsin.  It is familiar turf to me because that is where I finished psychiatric residency training.  I was impressed with the first two updates and have covered my experience at the 1st Annual and 2nd Annual Updates on this blog in the past.  During the introductory comments, the Department Chair Ned Kalin, MD paid tribute to John Greist, MD and James "Jeff" Jefferson, MD who ran the conference for 37 years before it was taken over by the Department of Psychiatry.

The absolute high point of day one was a discussion of schizophrenia by Daniel Weinberger, MD.  I had seen him speak before.  In talking with Dr. Kalin about the conference Dr. Weinberger gave a more research oriented talk the day before in the department that was focused on more science and neuroscience that I wish I had seen, but the lecture he gave to clinicians at this conference was outstanding.  From the introduction Dr. Weinberger had apparently left the NIMH and is now working as Director and CEO and  for the Lieber Institute for Brain Development.   They have quite a unique website and faculty.  The title of his discussion was Neurobiology of psychosis in the era of genetic medicine and he offered some unique perspectives that I doubt are available in many places.  It was useful to see a focus on schizophrenia in a conference of this nature and a focus on how at least some of the cutting edge research is looking at therapeutic modalities that are unique rather than the usual isomeric approach to drug discovery.  It was also refreshing to hear that there is optimism out there rather than the usual doom and gloom about the the "pipeline" from Big Pharma is running dry and there will be limited options for the future.

The Weinberger lecture had an interesting introduction that focused on brain imaging studies in psychiatric disorders.  It was well done, well argued, and based on his American J Psychiatry article that was published earlier this year (2).  He points out that while brain imaging has been the "centerpiece" of neuropsychiatric research that it is still fraught with technical difficulties.  In many of the articles that seem to make the lay press there is almost nothing that is not associated with brain changes.  He showed examples attempting to correlate viewing pornography with the frontostriatal network, increased gray matter volume secondary to lithium use, and other common artifacts and he concludes that most illness associated imaging findings are likely epiphenomenal.  To anyone trained as a chemist in their undergrad major who has experience with nuclear magnetic resonance (NMR) scanning of organic molecules a lot of this comes as no surprise.  To anyone used to reading decades of similar research (like quantitative EEG) and realizing that the pilot studies never panned out even after some were published in very prestigious journals this should also not come as a surprise.  Weinberger offered the technical explanations for why these issues occur and also some studies that seemed to be sound.

The bulk of his lecture was dedicated to the genetics of schizophrenia.  The opening slide not only contained a lot of information it was a tutorial in how to present information in PowerPoint format.  It was titled "The emerging genetics of schizophrenia".  In the upper left corner was a graphic from Gottesman's work with 11 bar graphs above the same axis showing risk for schizophrenia based on relationship to the index case.   Right below that was a table of Exome Sequencing: Rare Variants showing rare structural variants in schizophrenia with the title of a report to the right.  In the upper right hand corner was a Manhattan plot of 108 GWAS loci on all human chromosomes and the reference to the report in Nature.  It was a beautiful slide in terms of presentation and information content.

He went on to discuss genome-wide association studies (GWAS) and what they imply for the genetics of schizophrenia.  Inheritance of schizophrenia is widely considered to be polygenic and has been for some time.  He framed this issue as there being no psychiatric disorder gene and I thought that was a useful reframing because it speaks to studies that are looking at a very few point mutations associated with schizophrenia, and it is easy to think that this gene is the cause of schizophrenia rather than conferring risk for the disorder.  He demonstrated this with a risk profile score (RPR) for developing schizophrenia based on an additive count of all risk alleles.  In the example given the risk profile for the highest risk score had an odds ration of 15-20 to 1 for developing schizophrenia.  He went on to review the evidence for schizophrenia as a neurodevelopmental disorder.  That included some epidemiological data such as artificially imposed famine in China and the Netherlands and the subsequent increase in the incidence of schizophrenia in the respective birth cohorts (1).  He showed that de novo mutation in schizophrenia overlap with more traditional neurodevelopmental disorders like autism spectrum disorder and intellectual disability.  He showed that genes from all three disorders are overexpressed during the fetal period and this is a pattern seen in neurodevelopmental disorders.  

This was compelling stuff.  I come to this conference very year to get rejuvenated and it worked again this year.  The only regret I had was that time has just about run out for me.  I am no longer a young science major with an interest in human behavior and how the brain works.  I don't have time to go back and do a fellowship with Dr. Weinberger or a sleep fellowship or any number of other interesting things that I see at conferences.  I can understand the concepts,  teach them, and advise younger colleagues and residents on what is available and why this is a compelling field whenever I can.  I can also continue to get the word out that psychiatry is alive and well, that the best critics of psychiatry are trained as psychiatrists, and what passes for psychiatric criticism on blogs and in the press lacks a critical element called scholarship.  And as important - you don't have to be Kandel or Weinberger to be scholarly and apply what you know about the science to what you do every day as a psychiatrist.  Equally important - knowing the theory and what can and cannot be applied yet - is an important aspect of being a physician.

It was a good weekend.


George Dawson, MD, DFAPA


References:


1:  Schizophrenia and famine collection (original articles on the Dutch and Chinese famine are references number 39 and 59):

http://www.ncbi.nlm.nih.gov/sites/myncbi/1-MAvBcofi/collections/48942475/public/

2:  Weinberger DR, Radulescu E. Finding the Elusive Psychiatric "Lesion" With21st-Century Neuroanatomy: A Note of Caution. Am J Psychiatry. 2015 Aug 28:appiajp201515060753. [Epub ahead of print] PubMed PMID: 26315983.

DSM5 Dead on Arrival!

That's right.  The latest sensational blast on the fate of that darling of the media the DSM5 is that it is dead on arrival.  That recent proclamation is from the Neuroskeptic and it is based on the analysis of  criticism of DSM5 criteria for Generalized Anxiety Disorder (GAD).  OK - the original proclamation was "increasingly likely DOA".  I confess that at this point I have not read the original article by Starcevic, Portman, and Beck but the Neuroskeptic provides significant excerpts and analysis.

The broad criticism is that the category has been expanded and is therefore less specific.  The authors are concerned that this will lead to more inclusion and that will have "negative consequences."  The main concern is the "overmedicalization" of the worried and the dilution of clinical trails.  All this gnashing of the teeth leads me to wonder if anyone has actually read the Generalized Anxiety Disorder DSM5 criteria that is available on line.  The proposed new criteria, the old DSM-IV criteria and the rationale for the changes are readily observed.  The basic changes include a reduction on the time criteria for excessive worry from 6 months to three months, the elimination of criteria about not being able to control worry, and the elimination of 4/6 symptoms under criteria C (easy fatigue, difficulty concentrating, irritability and sleep disturbance).  A new section on associated behaviors including avoidance behavior a well known feature of anxiety disorders is included.  The remaining sections on impairment and differential diagnosis are about the same.  The GAD-7 is included as a severity measure although I note that the Pfizer copyright is not included.

So what about all of the criticism?  The "Rationale" tab is a good read on the DSM5 web site.  I can say that clinically non-experts are generally clueless about the DSM-IV features of anxiety especially irritability.  Most psychiatrists have a natural interest in irritability because we tend to see a lot of irritable people.  There has been some isolated work on irritability but it really has not produced much probably because it is another nonspecific symptoms that cuts across multiple categories like the authors apply to cognitive problems and pain.  So I will miss irritability but not much.  Psychiatrists have to deal with it whether we have a category for it or not and hence the need for a diagnostic formulation in addition to a DSM diagnosis (managed care time constraints permitting).

But like most things psychiatric - the worried masses rarely present to psychiatrists for treatment these days.   How likely is it that a busy primary care physician is going to review ANY DSM criteria for GAD?  How likely is it that a person with a substance abuse disorder is going to disclose those details to a primary care physician as a probable cause of their anxiety disorder?  How likely is it that benzodiazepines will be avoided as a first line treatment for any anxiety disorder?  In my experience as an addiction psychiatrist I would place the probability in all three questions to be very low.  It doesn't really matter if you use DSM-IV criteria or DSM5 criteria - the results are the same.

As far as "medicalization" goes, I am sure that somebody (probably on the Huffington Blog) will whip this into another rant about how the DSM5 enables psychiatrists to overdiagnose and overprescribe in our role as stooges for Big Pharma.  But who really has an interest in treating all anxiety like a medical problem?  I have previously posted John Greist's  single handed efforts in promoting psychotherapy and computerized psychotherapy for anxiety disorders even to the point of saying that the results are superior to pharmacotherapy.  In the meantime, what has the managed care cartel been doing?  Although their published guidelines appear to be nonexistent it would be difficult to not see the parallels between approaches that use the PHQ-9 to assess and treat depression and using the parallel instrument GAD-7 in a similar manner.  The problem with both approaches is that they are acontextual and the severity component cannot be adequately assessed.  The goal of managed care approaches to treat depression is clearly to get as many people on medications as possible and call that adequate treatment.  Why would the treatment of GAD be any different?

It should be obvious at this point that I am not too concerned about the DSM5, DSM-IV, or whatever diagnostic system somebody wants to use.  The DSM5 is clearly about rearranging criteria based on recent studies with the sole exception of including valid biological markers for the sleep disorders section.  Like many my speculation is that the ultimate information based approach to psychiatric disorders rests in genomics and refined epigenetic analysis and I look forward to that information being incorporated at some point along the way.

But let's get realistic about why the results of DSM technology are limited.  As it is with DSM-IV and as it will be with DSM5, clinicians are free to interpret and diagnose basically whatever they want.  Even with the vagaries of a DSM diagnosis, I doubt that the majority of primary care treatment hinges on a DSM diagnosis of any sort.  I also doubt that the dominant managed care approach to diagnosis and treatment of GAD depends on a psychiatric diagnosis or research based treatment.  It certainly excludes psychotherapy.  Trying to pin those serious deficiencies as well as overexposure to medication on the DSM and psychiatrists is folly.

George Dawson, MD, DFAPA


1: Gorman JM. Generalized anxiety disorders. Mod Probl Pharmacopsychiatry. 1987; 22: 127-40. PubMed PMID: 3299062.

Revolutionizing the Treatment of Anxiety and Depression

In a word - computers.

I had the good fortune of training with John Greist, MD  at the University of Wisconsin in the 1980s.  Interestingly, many people have the opinion that Dr. Greist is firmly in the camp of biomedical psychiatry.  He and his long time colleague James Jefferson, MD regularly give Door County symposia on the medical treatment of mood and anxiety disorders.  They are highly regarded for their scholarship and teaching ability.  If you haven't listened closely enough over the years, you might miss the fact that Dr. Greist has consistently pointed out the superiority of psychotherapy for various conditions and that  computerized versions of the same psychotherapy perform as well as seeing a therapist.

At a recent MPS meeting, Dr. Greist gave a presentation on computerized therapy.  He made a compelling argument for computerized psychotherapy based on a recent meta-analysis of effectiveness and a comparison of the cost effectiveness of developing moderately effective drugs compared to very cost effective and potentially more effective computerized psychotherapies.  He was an innovator in the field publishing some of the original research and designing some of the original software.  At this meeting he made a strong argument that the software is inexpensive, potentially as effective and more consistent than human therapists and for many conditions more effective than medication.

If there was any market value in the existing mental health field, Dr. Greist's concept would be disruptive.  It would potentially change the way that treatment is provided, especially treatment of anxiety and mood disorders.   Think about the way that treatment of these disorders is currently delivered.  Twenty percent of the adult population is at annual risk.  About 40 percent of that group seeks treatment primarily through primary care clinics.  Very few people see psychiatrists and very few people need to.  The standard of care for almost everyone else is taking a medication prescribed by a primary care clinic.  Many people are treated with benzodiazepines and sedative hypnotic medications that have no efficacy in anxiety or depression and they continue these medications on a chronic basis.  If psychotherapy is available it is two or three sessions of crisis intervention or supportive psychotherapy rather than research proven therapy for a specific disorder.

The lack of availability of psychotherapy in the health care system is another direct result of managed care and rationing.  Managing most of the anxiety and depression with medications and brief visits is ideal for the bean counters.  Outpatient clinics become an assembly line of 15 minute "med checks".   The only reality is a medication and whether that medication works and is tolerated.   An occasional manager may insist that the clinic double book patients to compensate for missed appointments or extra appointments to generate more revenue.

I noticed  today in an effort to send an e-mail to my internist that his primary care clinic offers e-mail consults on treating anxiety and depression for $40.  That is about what most psychiatrists get paid for a face-to face consultation.  I wonder if the $40 fee includes a description of the psychotherapies that might work better than medication?

Enter computerized psychotherapy.  Instead of waiting to get into a clinic that is based solely on medications, a person with anxiety and or depression accesses an Internet Clinic and proceeds through a number of self-guided and computerized cognitive behavioral therapy options.  There are options for preferences, combination therapies, and inadequate response to computerized therapy.  There is no need to travel to a clinic and there is no waiting.  The therapy is available on demand and for free. The cost of treating thousands of patients is trivial, basically limited to staff to maintain the web site, collect treatment data, analyze outcomes, and modify the software as necessary.

All of this has been a known possibility for about two decades.   Why isn't it happening?  Why is mental health treatment limited to medications when psychotherapy, even by a machine is superior in many cases?  Over those two decades we have seen unprecedented rationing of mental health services.  We have seen what used to be clinical decisions turned into business decisions.  The end result has not only been lower quality clinical care but a complete lack of innovation.  It is time for the pendulum to swing back in the right direction.    

George Dawson, MD, DFAPA

Andrews G, Cuijpers P, Craske MG, McEvoy P, Titov N (2010) Computer Therapy for the Anxiety and Depressive Disorders Is Effective, Acceptable and Practical Health Care: A Meta-Analysis. PLoS ONE 5(10): e13196. doi:10.1371/journal.pone.0013196