beyond a reflection…

Posted on Saturday 5 December 2015

Occam’s razor, also known as Ockham’s razor, and sometimes expressed in Latin as lex parsimoniae [the law of parsimony, economy or succinctness], is a principle that generally recommends selecting from among competing hypotheses the one that makes the fewest new assumptions. The strict translation is "Plurality is not to be posited without necessity."

In a reflection… I was talking about screening for depression  [waiting room screening], having encountered it as both patient and doctor at the beginning of the week of Thanksgiving. I ended by saying:
But I did decide that chasing this down would be worth the effort. It might take a while because this has obviously worked its way into government guidelines. So I’m going to table it for a time when I run out of other things to look into. In the meanwhile, I’m going to see how it worked its way into my clinic.
…thinking that with all the references that appeared when I looked into the why of things, there was a lot of reading between where I was and where I needed to be to understand why both my personal doctor and the clinic I work in were screening people for depression in the waiting room with questionnaires. So I started looking for meta-analyses and studies to figure it all out. Finding the answer to the why of this business of screening for depression had suddenly come into my life turned out to be much simpler than I thought it might be. Occam’s Razor proposes that the simplest explanation is usually right. And this is a prime example.

It takes just a couple of representative articles to summarize essence of this story. So let’s start with Canada. Back in 2005, the Canadian Task Force on Preventive Health Care had recommended depression screening. In 2012, there was a strong review article summarizing the available information in the interim. Like the Cochrane Review in a reflection…, they concluded that the Task Force should reconsider that recommendation:
by Brett D. Thombs, PhD, James C. Coyne, PhD, Pim Cuijpers, PhD, Peter de Jonge, PhD, Simon Gilbody, DPhil, John P. A. Ioannidis, MD DSc, Blair T. Johnson, PhD, Scott B. Patten, MD PhD, Erick H. Turner, MD, and Roy C. Ziegelstein, MD
Canadian Medical Association Journal. 2012 184[4]:413-418.

Conclusion:
The prevalence of depression and the availability of easy-to-use screening instruments make it tempting to endorse widespread screening for the disease. However, screening in primary care is a resource-intensive endeavour, does not yet show evidence of benefit and would have unintended negative effects for some patients. Evidence from one simulation study using Canadian national data found that the overall burden of depression could be reduced by providing more consistent treatment to reduce symptoms and prevent relapse among people with recurrent disorders, but not by increasing treatment through screening.

We hope that a rigorous review of current evidence will encourage the developers of future guidelines, including members of the Canadian Task Force on Preventive Health Care, to carefully consider their stance on screening for depression. We also hope that, consistent with the 2010 guidelines of the National Institute for Health and Clinical Excellence, such developers will conclude that evidence from well-conducted, randomized, controlled trials of the benefit of screening, in excess of its likely harms and costs, is needed before it can be recommended in primary care settings. Specifically, the benefits and harms of screening should be tested in a trial in which all patients identified as having depression should have access to the same integrated care for their condition, regardless of whether they are identified through screening in the intervention group or via physician recognition and referral in a control group. It is possible that such a trial would find that screening benefits patients to a degree that would justify the cost and the harms associated with the process. Until then, however, given the lack of evidence of benefit from screening and the concerns that we have described, it is not reasonable to simply assume that depression screening is good policy.

Key points
  • Screening for depression in primary care is recommended in the United States and Canada under certain conditions, but not in the United Kingdom.
  • No trials have found that patients who undergo screening have better outcomes than patients who do not when the same treatments are available to both groups.
  • Existing rates of treatment, high rates of false-positive results, small treatment effects and the poor quality of routine care may explain the lack of effect seen with screening.
  • Developers of future guidelines should require evidence of benefit from randomized controlled trials of screening, in excess of harms and costs, before recommending screening.
And that is, in fact, exactly what the Canadian Task Force on Preventive Health Care recommended the next year:
Canadian Task Force on Preventive Health Care
Canadian Medical Association Journal. 2013 185[9]:775-782.

Conclusion:
Our recommendations highlight the lack of evidence about the benefits and harms of routinely screening for depression in adults. In the absence of a demonstrated benefit of screening, and in consideration of the potential harms, we recommend not routinely screening for depression in primary care settings, either in adults at average risk or in those with characteristics that may increase their risk of depression. However, clinicians should be alert to the possibility of depression, especially in patients with characteristics that may increase their risk of depression, and should look for it when there are clinical clues, such as insomnia, low mood, anhedonia and suicidal thoughts.

Summary of recommendations for clinicians and policy-makers:
Recommendations on screening for depression in primary care settings are provided for people 18 years of age or older who present at a primary care setting with no apparent symptoms of depression. These recommendations do not apply to people with known depression, with a history of depression or who are receiving treatment for depression.
  • For adults at average risk of depression, we recommend not routinely screening for depression. [Weak recommendation; very-low-quality evidence]
  • For adults in subgroups of the population who may be at increased risk of depression, we recommend not routinely screening for depression. [Weak recommendation; very-low-quality evidence]
Their report is accompanied by a sensible Commentary:
by Roger C. Bland CM MB ChB, David L. Streiner PhD
Canadian Medical Association Journal. 2013 185[9]:753-754.

Key points
  • There is little evidence of sufficient quality to guide practitioners about what type of screening, if any, to use to detect depression in adults in primary care settings.
  • The number of false-positive screens with current assessment tools is too high, and the follow-up required to rule them out too time-consuming, to justify routine screening for depression in primary care practices.
  • If false-positive screens are not ruled out. patients are at increased risk of receiving the wrong diagnosis and inappropriate treatment.


Summary for Canada, US, UK
table from the Canadian Task Force on Preventive Health Care 2013

Screening is not recommended. In the US, it says don’t screen unless you are in a system that can offer full services to evaluate, treat, and follow-up the results.

So you might be asking at this point, "How this is a post to explain why 1boringoldman’s doctor and 1boringoldman’s clinic are both doing routine waiting room depression screening?" But that’s because you’re not thinking about the right thing – that Occam’s Razor proposes that the simplest explanation is usually right. And the answer is pretty simple. They’re doing waiting room depression screening because they’re getting paid to do waiting room depression screening.
Centers for Medicare & Medicaid Services [CMS]
Final Decision
October 14, 2011

The Centers for Medicare & Medicaid Services [CMS] has determined that the evidence is adequate to conclude that screening for depression in adults, which is recommended with a grade of B by the U.S. Preventive Services Task Force [USPSTF], is reasonable and necessary for the prevention or early detection of illness or disability and is appropriate for individuals entitled to benefits under Part A or enrolled under Part B.

Therefore CMS will cover annual screening for depression for Medicare beneficiaries in primary care settings that have staff-assisted depression care supports in place to assure accurate diagnosis, effective treatment and follow-up. For the purposes of this decision memorandum:
  • A primary care setting is defined as one in which there is provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs, developing a sustained partnership with patients, and practicing in the context of family and community. Emergency departments, inpatient hospital settings, ambulatory surgical centers, independent diagnostic testing facilities, skilled nursing facilities, inpatient rehabilitation facilities and hospice are not considered primary care settings under this definition.
  • At a minimum level, staff-assisted depression care supports consist of clinical staff [e.g., nurse, physician assistant] in the primary care setting who can advise physician of screening results and who can facilitate and coordinate referrals to mental health treatment.
Medicare [and Medicaid?] pays for waiting room screening for depression in adults once a year. The whole text is online to read. They review much of the same information in the other pieces above. They allow all sorts of screening instruments [the PHQ-9 seems the shortest]. They’ve softened the U.S. Preventive Services Task Force [USPSTF] recommendation that you only screen if you have full services to evaluate and deal with what you find out. All they require is that an assigned somebody inform the doctor of the results and then make a referral to some mental health asset. I suppose that could mean a local mental health center [or in the case of our clinic, me]. Best I can tell, they pay $17 and change for the yearly screening, and in the Medicare/Medicaid world where reimbursement is generally slim, every little bit counts. Having achieved something of a handle on the why of things, I think that’s enough for the moment and I’ll defer my editorial comments to the next post. I may not have the details totally right, but I think I’m hearing the right music. As they say, "Close enough for government work"…
  1.  
    1boringyoungman
    December 5, 2015 | 10:12 AM
     

    Please don’t forget the quality of care metrics. Where clinics will then have their bonuses linked to the percentage of the depressed indentified through screening who end up being placed on a SSRI a.k.a. “treated.”

  2.  
    James O'Brien, M.D.
    December 5, 2015 | 10:14 AM
     

    They’re forcing physicians to reduce the time of an encounter yet demanding that they screen for dozens of issues unrelated to the chief complaint.

    It’s Orwellian and mind-numbingly stupid.

    Physicians are like the Jonathan Pryce character in Brazil (perhaps the most underrated and relevant movie of the past thirty years). They are barely holding it together but it can’t last.

  3.  
    1boringyoungman
    December 5, 2015 | 11:14 AM
     

    It might be worth asking your medical director of the impact on your clinic for not doing the screening goes beyond the $17 cost. Would lay odds that now, or in the near future, failing to adequately “integrate” in mental health care will come at a greater cost than that to the clinic. Very few people are breaking down what all this will look like in practice.

  4.  
    Oneboringoldwoman
    December 5, 2015 | 1:33 PM
     

    I’m sure it will eventually lead to something similar to the already existing AFIX system used to track certain activities of doctors – where a doctor will get a knock on the door from someone from the government for a seemingly random check to determine why their numbers are not in line with the government/pharmaceutical goals, and then a subsequent mandatory intervention and re- education about how to increase the numbers if they want to avoid further visits from government representatives or medical board personnel. In the case of psychiatry, of course, it will pertain to the numbers of those treated with just a different kind of pharmaceutical product then that which AFIX currently tracks. Maybe there might even be government funded”bonuses” like with AFIX, so when a psychiatrist reaches a certain percentage of coverage (dispensing of the product) in his/her practice, he makes a bunch more money.

  5.  
    1boringyoungman
    December 5, 2015 | 5:33 PM
     

    http://www.cdc.gov/vaccines/programs/afix/components.html
    There is a much much better case to be made for attempting to boost immunization %s to achieve herd immunity. But yes, this approach can be expanded across a variety of metrics in much more questionable ways. A metric can be a good straightforward gauge (eg % population immunized) or not (as we are seeing in early childhood education eg https://www.washingtonpost.com/news/answer-sheet/wp/2015/11/24/how-twisted-early-childhood-education-has-become-from-a-child-development-expert/ ).
    Perhaps the greatest failure of the modern DSMs is the MDD category. First it was yoked to Pharma. Now yoked to a public health model.
    Dr. O’Brien’s reference to Brazil seems apt. The drive and instantiation around much of this is banal.

  6.  
    James O'Brien, M.D.
    December 5, 2015 | 7:08 PM
     
  7.  
    Catalyzt
    December 5, 2015 | 8:12 PM
     

    Early intervention is great when it makes sense. My mom did groundbreaking work on early intervention in learning disabilities, but that was developmental, completely different animal, obviously.

    The Freudians kind of argued for early intervention, if memory serves, though they weren’t very big on screening instruments, thank God. The general idea seemed to be that analysis worked better for adolescents, or it was faster, anyway, and that was certainly my experience. I knocked it out in four years or so, but adults seemed to get stuck in it forEVer.

    It certainly worked like a charm for knowing what I wanted to do with my life and steering clear of existential torpor, though I don’t think it did anything to change my mood or decrease anxiety or whatever it is that people think psychotherapy should accomplish these days.

    I had no idea the PHQ9 came from Pfizer.

    Sorry, I’m rambling.

  8.  
    December 5, 2015 | 8:40 PM
     

  9.  
    Tom
    December 5, 2015 | 10:29 PM
     

    I apologize for being well versed in psychometrics. I deal with the issue of self-report psychological inventories in clinical and forensic settings. Let me just say that no self-report instrument is worth anything unless you have ancillary measures of response style– like how do we know the results are believable or pass the smell test? Is the respondent even paying attention to the items? Is the respondent answering in a consistent manner? Is the respondent exaggerating or minimizing distress? Tests like the MMPI, PAI and MCMI have scales to assess such matters. Even the Rorschach has measures of validity. These screening measures (PHQ9 whatever) provide no information about these matters.

  10.  
    Catalyzt
    December 5, 2015 | 11:25 PM
     

    @ Mickey — our copy of the PHQ9 at the Center had been Xeroxed so many times, you could barely read the questions, let alone the fine print. I doubt most folks who worked in MHS had any idea there was any commercial interest involved in its development.

    @ Tom — Exactly. Not being at all well-versed in psychometrics, I just know how much I appreciate it when the intake clinician physically writes notes on the form itself (as well as putting it in some notes section somewhere in the EMR) e.g.: “Responses incongruent with affect and hx” or “Client admitted to being extremely intoxicated at end of interview,” etc.

    As Dr. O’Brien and others have noted, the issue is time, whether in primary care or elsewhere, and lost time can mean increased risk.

    Aside from that, sometimes just remembering to do fill out the instrument (along with the million other things you have to remember for grant or DMH compliance, because the list of things is constantly changing) is more irritating than actually doing it. One of the most absurd situations interns ran into was when they forgot to administer some instrument– usually QIDS– at six months or termination or whatever, and would wind up calling the client a year later and saying, “I know this sounds insane, but please fill this out as if it were August of 2014 and send it back to me.”

  11.  
    James O'Brien, M.D.
    December 6, 2015 | 1:12 AM
     

    Never a need to apologize for interest in psychometrics…there needs to be more…for now they are what biomarkers hope to be. Tom is absolutely right about validity tests and the need to measure inconsistent responses, endorsement of unusual items and denial of ordinary faults, which place self report inventories in context. Even the self report inventories better than the PHQ-9, which means every other one.

    It took three people and a grant to do that? All he did was basically paraphrase DSM. They should have asked me, I’d have ready for them in an hour.

  12.  
    James O'Brien, M.D.
    December 6, 2015 | 11:45 AM
     

    Another problem with this test…and this applies to the Beck Depression inventory and DSM as well…obviously #1 and #9 are a much bigger deal and more specific to BIG D depression than #3 and #7 but are not weighted…

    I read the original papers on the “validity” of the PHQ-9 and basically it says this…if a patient endorses the 9 symptoms of major depression to a certain degree then it correlates with the diagnosis of major depression. Are you kidding me? That’s a tautology. And the authors of article has absolutely no idea what construct validity really is….because implicit is their assumption that DSM depression has construct validity.

    I could create a nine point diagnosis for “Adele fan” then ask people in a clinic the same nine questions. Viola! That survey would be highly predictive of who meets the Adele fan syndrome. It tells you nothing about whether Adele fan is a mental illness or a normal variant.

    Psychiatry is so dumbed down right now it laughable if weren’t so pathetic.

  13.  
    James O'Brien, M.D.
    December 6, 2015 | 1:24 PM
     

    Hey, I just came up with a new “test” for primary care doctors to use…I call it the
    MED-3. Here it is:

    1. Is it difficult to attain adequate an adequate erection?

    a. no difficulty
    b. some difficulty
    c. very difficult
    d. extremely difficult

    2. Do this cause distress or interpersonal difficulty?

    a. no difficulty
    b. some difficulty
    c. much difficulty
    d. extreme difficulty.

    3. Is this not better accounted for by another Axis 1 disorder, a substance use disorder or medical condition?

    a. explained by those factors
    b. not explained by those factors

    AMAZINGLY, the results of this test correlate with DSM diagnosis of Male Erectile Disorder!!!

    Where’s my honoraria from Cialis, Levitra and Viagra? My residents and interns are requesting further field trials and the Four Seasons in St. Kitts isn’t cheap.

    BTW this should be given to all primary care patients regardless of sex. I want my Addiya grant too. Field trials at Magic Mike’s.

    And I did this without two coauthors. In about 10 minutes.

  14.  
    Bernard Carroll
    December 6, 2015 | 2:51 PM
     

    Very good, James… be sure now to copyright it. 🙂

  15.  
    1boringyoungman
    December 6, 2015 | 3:35 PM
     

    “because implicit is their assumption that DSM depression has construct validity.” This is an important point.

  16.  
    James O'Brien, M.D.
    December 6, 2015 | 4:55 PM
     

    At this point the MTV level vapidity actually bothers me more than the corruption.

    During the analyst-dominated era, thought leaders were often wrong, but at least interesting and thought provoking.

  17.  
    Bernard Carroll
    December 6, 2015 | 4:56 PM
     

    PHQ-9 generates shadows on the wall of Plato’s cave. PHQ = Psychiatry’s Happiness Quantifier.

  18.  
    Altostrata
    December 7, 2015 | 7:43 PM
     

    How is it that mammograms and prostate cancer screening get such minute, continuous cost-benefit analyses to minimize the harm of false positives, but “screening” for “depression” does not?

  19.  
    James O'Brien, M.D.
    December 9, 2015 | 10:30 AM
     

    My screening test makes more sense than this ten point screening for preschool ADHD:

    http://www.psychiatrictimes.com/adhd/assessing-adhd-preschool-children

    Do child psychiatry researchers have kids?

    This post was developed and funded by a grant from Chuck E Cheese.

  20.  
    1boringyoungman
    December 11, 2015 | 2:46 PM
     

    “In addition to routine screenings, the newly published data describe how reliably physicians provide follow-up care for depression…”
    http://www.modernhealthcare.com/article/20151210/NEWS/151219991

Sorry, the comment form is closed at this time.