The Limitations of Checklists in Assessing Mental HealthFriday, July 13, 2012
An assessment of mental health (the first ‘M’ of obesity) should be part of every assessment for obesity. Not only can virtually all mental health problems (from mild to severe) promote weight gain, but they can and, in virtually all severe cases, will, present significant barriers to weight management.
Unfortunately, most practitioners, who are called upon to give weight management advice, have little to no formal training in diagnosing, let alone dealing with, mental health disorders.
This is when they sooner or later fallback on checklists – often self-administered – to screen for the presence of mental illness.
In a ‘perspective’ piece recently published in the New England Journal of Medicine, Johns Hopkins University’s Paul McHugh and Phillip Slavney, discuss the limitations of overly relying on checklists for mental health assessments.
As they explain, the use of checklists to assess the presence of ‘symptoms’, proliferated in psychiatric practice following the introduction of the Diagnostic and Statistical Manual (SDM), which introduced a much needed ‘standardisation’ in the assessment of mental health disorders.
As a downside however, simply counting or checking off symptoms on a list has resulted in a shift towards ‘phenomenology’ rather than to consideration of etiological differences in clinical symptoms that may appear similar both to the patient and to an observer.
Thus, for example, simply counting off symptoms, may lead to the ‘diagnosis’ of depression in someone who is bereaved, has classic melancholia or is simply ‘demoralised’ by circumstances. (I would add untreated sleep apnea to the list of potential confounders)
As the authors note:
“The mixing of similar-appearing patients who have conditions that are distinct in nature probably explains why the use of this diagnostic category expanded over time and suggests why the effectiveness of antidepressant medications given to people with a diagnosis of major depression has, of late, been questioned.”
“This tendency to blur natural distinctions may explain why other DSM diagnoses – such as post-traumatic stress disorder (PTSD) and attention deficit disorder – have been overused if not abused.”
Indeed, as the authors explain, the introduction of the DSM with,
“Its emphasis on manifestations persuaded psychiatrists to replace the thorough ‘bottom-up’ method of diagnosis, which was based on a detailed life history, painstaking examination of mental status, and corroboration from third-party informants, with the cursory ‘top-down’ method that relied on symptom checklists.”
Thus, although “checklist diagnoses” cost less time and money, “they fail woefully to correspond with diagnoses derived from comprehensive assessments.”
This not only results in over and mis-diagnoses but can also be misused for ‘marketing’ certain ‘diagnoses’ by clinicians, industry, and other interest groups.
To remedy this phenomenological approach, the authors propose a more ‘etiological’ approach that distinguishes between brain diseases (e.g. schizophrenia), personality dimensions (e.g. obsessive-compulsive personality), motivated behaviours (e.g. alcoholism), and life encounters (e.g. bereavement, trauma) – recognising that often several of these ‘etiological’ factors may overlap.
Despite such overlap, the authors suggest that such an ‘etiological’ grouping of disorders would encourage a more analytical approach to assessments and treatments.
Regular readers may discern similarities between this approach and the ‘etiological’ approach that I have previously recommended for the ‘differential’ diagnosis of obesity.
Thus, although using checklists will certainly continue to be a simple way to screen individuals, to rely on them for making a diagnosis or to initiate treatments based on the number of ticked boxes would be poor medical practice.
This should not discourage us from using symptom checklists in our practice – it should, however, encourage us to delve deeper into the underlying causes and impacts of these symptoms before pronouncing a diagnosis or ‘labeling’ someone with a mental health disorder.
p.s. Click here to download a copy of the Toronto Charter on Obesity and Mental Health
McHugh PR, & Slavney PR (2012). Mental illness–comprehensive evaluation or checklist? The New England journal of medicine, 366 (20), 1853-5 PMID: 22591291
Saturday, July 14, 2012
This quote from your article today (re: “Limitations of Checklists in Assessing Mental Health”) presents a rather ominous view of diagnostic practices that could result in harmful consequences–not merely for diagnosed individuals (a serious concern) but for populations, such as those categorized as “obese”:
“Thus, although ‘checklist diagnoses’ cost less time and money, ‘they fail woefully to correspond with diagnoses derived from comprehensive assessments.’ This not only results in over and mis-diagnoses but can also be misused for ‘marketing’ certain ‘diagnoses’ by clinicians, industry, and other interest groups.”
I wonder if the “checklist diagnosis” was used for the study you cited, Dr. Sharma, on June 27, 2012, which claimed “… that slightly over one-quarter of obese individuals may suffer from varying degrees of borderline personality disorder.”
That percentage (greater than 25%), whether valid or invalid, could greatly increase the stigma associated with obesity (if reported in mass media sources) because Borderline Personality Disorder carries a significant diagnostic stigma. One study found that patients with mental illness would prefer to publicly admit to having Bipolar Illness than to be diagnosed in private with Borderline Personality Disorder (BPD) because the stigma that accompanies the latter diagnosis is perceived as so very destructive.
It would be helpful to be able to read the study that resulted with such a high percentage (correlating BPD and obesity), particularly the methodology section. A link would be great. Thanks.